arnauad3 commited on
Commit
65de7c4
·
1 Parent(s): b72c98c

November Release

Browse files
app.py CHANGED
@@ -15,13 +15,17 @@ from static.html_content import (
15
  LC_FOOTNOTE_HTML,
16
  NAV_BUTTONS_HTML,
17
  REFERENCES_HTML,
 
18
  )
19
  from style.css_html_js import custom_css
20
 
21
 
22
- def make_leaderboard_tab(state: Simulator, name: str):
23
- """Create a leaderboard tab with the given name and state."""
24
  with gr.Tab(name):
 
 
 
25
  # 1st row filters (select task, benchmark and sim)
26
  with gr.Row(equal_height=True):
27
  with gr.Column(scale=4):
@@ -142,7 +146,7 @@ with gr.Blocks(css=custom_css, theme=gr.themes.Default(primary_hue=colors.emeral
142
  model_type_dropdown_other,
143
  params_slider_other,
144
  leaderboard_other,
145
- ) = make_leaderboard_tab(state, name_other)
146
 
147
  # all plots using Plotly
148
  with gr.Tab("Plot View"):
 
15
  LC_FOOTNOTE_HTML,
16
  NAV_BUTTONS_HTML,
17
  REFERENCES_HTML,
18
+ OTHER_MODELS_HTML,
19
  )
20
  from style.css_html_js import custom_css
21
 
22
 
23
+ def make_leaderboard_tab(state: Simulator, name: str, html: str=None):
24
+ """ Create a leaderboard tab with the given name and state """
25
  with gr.Tab(name):
26
+ """" Add text box """
27
+ if html:
28
+ gr.HTML(html)
29
  # 1st row filters (select task, benchmark and sim)
30
  with gr.Row(equal_height=True):
31
  with gr.Column(scale=4):
 
146
  model_type_dropdown_other,
147
  params_slider_other,
148
  leaderboard_other,
149
+ ) = make_leaderboard_tab(state, name_other, html=OTHER_MODELS_HTML)
150
 
151
  # all plots using Plotly
152
  with gr.Tab("Plot View"):
config/constants.py CHANGED
@@ -6,15 +6,18 @@ VERILATOR_AGG = f"{RESULTS_DIR}/aggregated_scores_verilator.csv"
6
 
7
 
8
  DISCARDED_MODELS = {
 
 
 
 
 
 
 
 
 
9
 
10
  }
11
 
12
- """
13
- "DeepSeek R1": "10/10/2025",
14
- "QwenCoder 2.5 7B": "11/10/2025",
15
- "RTLCoder Mistral": "14/10/2025"
16
- """
17
-
18
  TASKS = ["Spec-to-RTL", "Code Completion", "Line Completion †"]
19
  S2R_BENCHMARKS = ["VerilogEval S2R", "RTLLM"]
20
  CC_BENCHMARKS = ["VerilogEval MC", "VeriGen"]
 
6
 
7
 
8
  DISCARDED_MODELS = {
9
+ "CodeV-CL-7B": "23/10/2025",
10
+ "CodeV-DS-6.7B": "23/10/2025",
11
+ "CodeV R1 Distill Qwen 7B": "23/10/2025",
12
+ "QwenCoder 2.5 7B": "23/10/2025",
13
+ "DeepSeek Coder 33B": "23/10/2025",
14
+ "RTLCoder Mistral": "23/10/2025",
15
+ "CodeLlama 70B": "23/10/2025",
16
+ "Llama 3.(1-3) 70B": "23/10/2025",
17
+ "DeepCoder 14B": "23/10/2025",
18
 
19
  }
20
 
 
 
 
 
 
 
21
  TASKS = ["Spec-to-RTL", "Code Completion", "Line Completion †"]
22
  S2R_BENCHMARKS = ["VerilogEval S2R", "RTLLM"]
23
  CC_BENCHMARKS = ["VerilogEval MC", "VeriGen"]
config/model_metadata.py CHANGED
@@ -112,4 +112,13 @@ MODELS = {
112
  "Seed-OSS-36B": ModelMetadata(
113
  "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct", 36.2, "General", "V3", "Reasoning"
114
  ),
 
 
 
 
 
 
 
 
 
115
  }
 
112
  "Seed-OSS-36B": ModelMetadata(
113
  "https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Instruct", 36.2, "General", "V3", "Reasoning"
114
  ),
115
+ "Qwen3-8B": ModelMetadata(
116
+ "https://huggingface.co/Qwen/Qwen3-8B", 8.2, "General", "V3", "Reasoning"
117
+ ),
118
+ "Hermes-4-14B": ModelMetadata(
119
+ "https://huggingface.co/NousResearch/Hermes-4-14B", 14, "General", "V3", "Dense"
120
+ ),
121
+ "Hermes-4-14B-Reasoning": ModelMetadata(
122
+ "https://huggingface.co/NousResearch/Hermes-4-14B", 14, "General", "V3", "Reasoning"
123
+ ),
124
  }
results/aggregated_scores_icarus.csv CHANGED
@@ -14,6 +14,9 @@ Magistral Small 2506,40.82,22.62,40.96,19.96,40.37,46.88
14
  gpt-oss-20b,63.7,66.48,67.41,68.44,51.46,48.58
15
  StarChat2 15B v0.1,38.76,38.19,36.68,35.58,45.61,61.96
16
  DeepSeek R1 Distill Qwen 14B,23.14,24.57,24.94,24.3,17.22,27.07
 
 
 
17
  CodeLlama 70B,33.04,33.68,32.2,32.27,35.81,46.56
18
  DeepSeek Coder 33B,27.03,38.47,22.65,37.64,41.47,46.01
19
  QwenCoder 2.5 32B,44.02,45.08,43.68,44.05,45.15,54.52
 
14
  gpt-oss-20b,63.7,66.48,67.41,68.44,51.46,48.58
15
  StarChat2 15B v0.1,38.76,38.19,36.68,35.58,45.61,61.96
16
  DeepSeek R1 Distill Qwen 14B,23.14,24.57,24.94,24.3,17.22,27.07
17
+ Hermes-4-14B-Reasoning,50.33,37.25,54.75,35.24,35.74,55.54
18
+ Hermes-4-14B,42.77,27.96,41.15,26.11,48.1,44.83
19
+ Qwen3-8B,45.1,48.16,48.72,49.24,33.17,38.32
20
  CodeLlama 70B,33.04,33.68,32.2,32.27,35.81,46.56
21
  DeepSeek Coder 33B,27.03,38.47,22.65,37.64,41.47,46.01
22
  QwenCoder 2.5 32B,44.02,45.08,43.68,44.05,45.15,54.52
results/aggregated_scores_verilator.csv CHANGED
@@ -14,6 +14,9 @@ Magistral Small 2506,41.02,23.48,41.25,20.36,40.27,51.89
14
  gpt-oss-20b,63.19,65.93,66.45,67.71,52.45,49.67
15
  StarChat2 15B v0.1,40.2,38.43,37.97,35.85,47.54,61.91
16
  DeepSeek R1 Distill Qwen 14B,22.93,24.82,24.44,24.43,17.96,28.33
 
 
 
17
  CodeLlama 70B,34.55,35.64,33.58,33.89,37.75,51.62
18
  DeepSeek Coder 33B,27.93,39.42,23.14,38.14,43.71,51.12
19
  QwenCoder 2.5 32B,45.72,47.29,45.31,46.49,47.07,54.62
 
14
  gpt-oss-20b,63.19,65.93,66.45,67.71,52.45,49.67
15
  StarChat2 15B v0.1,40.2,38.43,37.97,35.85,47.54,61.91
16
  DeepSeek R1 Distill Qwen 14B,22.93,24.82,24.44,24.43,17.96,28.33
17
+ Hermes-4-14B-Reasoning,50.57,38.19,55.07,36.12,35.74,57.11
18
+ Hermes-4-14B,44.05,28.99,42.69,26.97,48.52,47.41
19
+ Qwen3-8B,46.23,48.83,49.8,49.98,34.44,38.32
20
  CodeLlama 70B,34.55,35.64,33.58,33.89,37.75,51.62
21
  DeepSeek Coder 33B,27.93,39.42,23.14,38.14,43.71,51.12
22
  QwenCoder 2.5 32B,45.72,47.29,45.31,46.49,47.07,54.62
results/results_icarus.json CHANGED
@@ -4124,6 +4124,831 @@
4124
  "Release": "V1",
4125
  "Thinking": "Reasoning"
4126
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4127
  {
4128
  "Model": "CodeLlama 70B",
4129
  "Model Type": "Coding",
 
4124
  "Release": "V1",
4125
  "Thinking": "Reasoning"
4126
  },
4127
+ {
4128
+ "Model": "Hermes-4-14B-Reasoning",
4129
+ "Model Type": "General",
4130
+ "Benchmark": "VerilogEval S2R",
4131
+ "Task": "Syntax (STX)",
4132
+ "Result": 79.74,
4133
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4134
+ "Params": 14,
4135
+ "Release": "V3",
4136
+ "Thinking": "Reasoning"
4137
+ },
4138
+ {
4139
+ "Model": "Hermes-4-14B-Reasoning",
4140
+ "Model Type": "General",
4141
+ "Benchmark": "RTLLM",
4142
+ "Task": "Syntax (STX)",
4143
+ "Result": 76.6,
4144
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4145
+ "Params": 14,
4146
+ "Release": "V3",
4147
+ "Thinking": "Reasoning"
4148
+ },
4149
+ {
4150
+ "Model": "Hermes-4-14B-Reasoning",
4151
+ "Model Type": "General",
4152
+ "Benchmark": "VerilogEval S2R",
4153
+ "Task": "Functionality (FNC)",
4154
+ "Result": 55.9,
4155
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4156
+ "Params": 14,
4157
+ "Release": "V3",
4158
+ "Thinking": "Reasoning"
4159
+ },
4160
+ {
4161
+ "Model": "Hermes-4-14B-Reasoning",
4162
+ "Model Type": "General",
4163
+ "Benchmark": "RTLLM",
4164
+ "Task": "Functionality (FNC)",
4165
+ "Result": 39.57,
4166
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4167
+ "Params": 14,
4168
+ "Release": "V3",
4169
+ "Thinking": "Reasoning"
4170
+ },
4171
+ {
4172
+ "Model": "Hermes-4-14B-Reasoning",
4173
+ "Model Type": "General",
4174
+ "Benchmark": "VerilogEval S2R",
4175
+ "Task": "Synthesis (SYN)",
4176
+ "Result": 55.64,
4177
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4178
+ "Params": 14,
4179
+ "Release": "V3",
4180
+ "Thinking": "Reasoning"
4181
+ },
4182
+ {
4183
+ "Model": "Hermes-4-14B-Reasoning",
4184
+ "Model Type": "General",
4185
+ "Benchmark": "RTLLM",
4186
+ "Task": "Synthesis (SYN)",
4187
+ "Result": 34.47,
4188
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4189
+ "Params": 14,
4190
+ "Release": "V3",
4191
+ "Thinking": "Reasoning"
4192
+ },
4193
+ {
4194
+ "Model": "Hermes-4-14B-Reasoning",
4195
+ "Model Type": "General",
4196
+ "Benchmark": "VerilogEval S2R",
4197
+ "Task": "Power",
4198
+ "Result": 55.34,
4199
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4200
+ "Params": 14,
4201
+ "Release": "V3",
4202
+ "Thinking": "Reasoning"
4203
+ },
4204
+ {
4205
+ "Model": "Hermes-4-14B-Reasoning",
4206
+ "Model Type": "General",
4207
+ "Benchmark": "RTLLM",
4208
+ "Task": "Power",
4209
+ "Result": 38.35,
4210
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4211
+ "Params": 14,
4212
+ "Release": "V3",
4213
+ "Thinking": "Reasoning"
4214
+ },
4215
+ {
4216
+ "Model": "Hermes-4-14B-Reasoning",
4217
+ "Model Type": "General",
4218
+ "Benchmark": "VerilogEval S2R",
4219
+ "Task": "Performance",
4220
+ "Result": 53.73,
4221
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4222
+ "Params": 14,
4223
+ "Release": "V3",
4224
+ "Thinking": "Reasoning"
4225
+ },
4226
+ {
4227
+ "Model": "Hermes-4-14B-Reasoning",
4228
+ "Model Type": "General",
4229
+ "Benchmark": "RTLLM",
4230
+ "Task": "Performance",
4231
+ "Result": 32.73,
4232
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4233
+ "Params": 14,
4234
+ "Release": "V3",
4235
+ "Thinking": "Reasoning"
4236
+ },
4237
+ {
4238
+ "Model": "Hermes-4-14B-Reasoning",
4239
+ "Model Type": "General",
4240
+ "Benchmark": "VerilogEval S2R",
4241
+ "Task": "Area",
4242
+ "Result": 55.17,
4243
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4244
+ "Params": 14,
4245
+ "Release": "V3",
4246
+ "Thinking": "Reasoning"
4247
+ },
4248
+ {
4249
+ "Model": "Hermes-4-14B-Reasoning",
4250
+ "Model Type": "General",
4251
+ "Benchmark": "RTLLM",
4252
+ "Task": "Area",
4253
+ "Result": 36.13,
4254
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4255
+ "Params": 14,
4256
+ "Release": "V3",
4257
+ "Thinking": "Reasoning"
4258
+ },
4259
+ {
4260
+ "Model": "Hermes-4-14B-Reasoning",
4261
+ "Model Type": "General",
4262
+ "Benchmark": "RTL-Repo",
4263
+ "Task": "Exact Matching (EM)",
4264
+ "Result": -1.0,
4265
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4266
+ "Params": 14,
4267
+ "Release": "V3",
4268
+ "Thinking": "Reasoning"
4269
+ },
4270
+ {
4271
+ "Model": "Hermes-4-14B-Reasoning",
4272
+ "Model Type": "General",
4273
+ "Benchmark": "VerilogEval MC",
4274
+ "Task": "Syntax (STX)",
4275
+ "Result": 51.15,
4276
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4277
+ "Params": 14,
4278
+ "Release": "V3",
4279
+ "Thinking": "Reasoning"
4280
+ },
4281
+ {
4282
+ "Model": "Hermes-4-14B-Reasoning",
4283
+ "Model Type": "General",
4284
+ "Benchmark": "VeriGen",
4285
+ "Task": "Syntax (STX)",
4286
+ "Result": 74.12,
4287
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4288
+ "Params": 14,
4289
+ "Release": "V3",
4290
+ "Thinking": "Reasoning"
4291
+ },
4292
+ {
4293
+ "Model": "Hermes-4-14B-Reasoning",
4294
+ "Model Type": "General",
4295
+ "Benchmark": "VerilogEval MC",
4296
+ "Task": "Functionality (FNC)",
4297
+ "Result": 36.54,
4298
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4299
+ "Params": 14,
4300
+ "Release": "V3",
4301
+ "Thinking": "Reasoning"
4302
+ },
4303
+ {
4304
+ "Model": "Hermes-4-14B-Reasoning",
4305
+ "Model Type": "General",
4306
+ "Benchmark": "VeriGen",
4307
+ "Task": "Functionality (FNC)",
4308
+ "Result": 56.47,
4309
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4310
+ "Params": 14,
4311
+ "Release": "V3",
4312
+ "Thinking": "Reasoning"
4313
+ },
4314
+ {
4315
+ "Model": "Hermes-4-14B-Reasoning",
4316
+ "Model Type": "General",
4317
+ "Benchmark": "VerilogEval MC",
4318
+ "Task": "Synthesis (SYN)",
4319
+ "Result": 36.28,
4320
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4321
+ "Params": 14,
4322
+ "Release": "V3",
4323
+ "Thinking": "Reasoning"
4324
+ },
4325
+ {
4326
+ "Model": "Hermes-4-14B-Reasoning",
4327
+ "Model Type": "General",
4328
+ "Benchmark": "VeriGen",
4329
+ "Task": "Synthesis (SYN)",
4330
+ "Result": 56.47,
4331
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4332
+ "Params": 14,
4333
+ "Release": "V3",
4334
+ "Thinking": "Reasoning"
4335
+ },
4336
+ {
4337
+ "Model": "Hermes-4-14B-Reasoning",
4338
+ "Model Type": "General",
4339
+ "Benchmark": "VerilogEval MC",
4340
+ "Task": "Power",
4341
+ "Result": 35.26,
4342
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4343
+ "Params": 14,
4344
+ "Release": "V3",
4345
+ "Thinking": "Reasoning"
4346
+ },
4347
+ {
4348
+ "Model": "Hermes-4-14B-Reasoning",
4349
+ "Model Type": "General",
4350
+ "Benchmark": "VeriGen",
4351
+ "Task": "Power",
4352
+ "Result": 52.97,
4353
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4354
+ "Params": 14,
4355
+ "Release": "V3",
4356
+ "Thinking": "Reasoning"
4357
+ },
4358
+ {
4359
+ "Model": "Hermes-4-14B-Reasoning",
4360
+ "Model Type": "General",
4361
+ "Benchmark": "VerilogEval MC",
4362
+ "Task": "Performance",
4363
+ "Result": 35.52,
4364
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4365
+ "Params": 14,
4366
+ "Release": "V3",
4367
+ "Thinking": "Reasoning"
4368
+ },
4369
+ {
4370
+ "Model": "Hermes-4-14B-Reasoning",
4371
+ "Model Type": "General",
4372
+ "Benchmark": "VeriGen",
4373
+ "Task": "Performance",
4374
+ "Result": 56.22,
4375
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4376
+ "Params": 14,
4377
+ "Release": "V3",
4378
+ "Thinking": "Reasoning"
4379
+ },
4380
+ {
4381
+ "Model": "Hermes-4-14B-Reasoning",
4382
+ "Model Type": "General",
4383
+ "Benchmark": "VerilogEval MC",
4384
+ "Task": "Area",
4385
+ "Result": 34.94,
4386
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4387
+ "Params": 14,
4388
+ "Release": "V3",
4389
+ "Thinking": "Reasoning"
4390
+ },
4391
+ {
4392
+ "Model": "Hermes-4-14B-Reasoning",
4393
+ "Model Type": "General",
4394
+ "Benchmark": "VeriGen",
4395
+ "Task": "Area",
4396
+ "Result": 57.44,
4397
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4398
+ "Params": 14,
4399
+ "Release": "V3",
4400
+ "Thinking": "Reasoning"
4401
+ },
4402
+ {
4403
+ "Model": "Hermes-4-14B",
4404
+ "Model Type": "General",
4405
+ "Benchmark": "VerilogEval S2R",
4406
+ "Task": "Syntax (STX)",
4407
+ "Result": 72.18,
4408
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4409
+ "Params": 14,
4410
+ "Release": "V3",
4411
+ "Thinking": "Dense"
4412
+ },
4413
+ {
4414
+ "Model": "Hermes-4-14B",
4415
+ "Model Type": "General",
4416
+ "Benchmark": "RTLLM",
4417
+ "Task": "Syntax (STX)",
4418
+ "Result": 80.85,
4419
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4420
+ "Params": 14,
4421
+ "Release": "V3",
4422
+ "Thinking": "Dense"
4423
+ },
4424
+ {
4425
+ "Model": "Hermes-4-14B",
4426
+ "Model Type": "General",
4427
+ "Benchmark": "VerilogEval S2R",
4428
+ "Task": "Functionality (FNC)",
4429
+ "Result": 42.31,
4430
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4431
+ "Params": 14,
4432
+ "Release": "V3",
4433
+ "Thinking": "Dense"
4434
+ },
4435
+ {
4436
+ "Model": "Hermes-4-14B",
4437
+ "Model Type": "General",
4438
+ "Benchmark": "RTLLM",
4439
+ "Task": "Functionality (FNC)",
4440
+ "Result": 48.09,
4441
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4442
+ "Params": 14,
4443
+ "Release": "V3",
4444
+ "Thinking": "Dense"
4445
+ },
4446
+ {
4447
+ "Model": "Hermes-4-14B",
4448
+ "Model Type": "General",
4449
+ "Benchmark": "VerilogEval S2R",
4450
+ "Task": "Synthesis (SYN)",
4451
+ "Result": 42.31,
4452
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4453
+ "Params": 14,
4454
+ "Release": "V3",
4455
+ "Thinking": "Dense"
4456
+ },
4457
+ {
4458
+ "Model": "Hermes-4-14B",
4459
+ "Model Type": "General",
4460
+ "Benchmark": "RTLLM",
4461
+ "Task": "Synthesis (SYN)",
4462
+ "Result": 45.53,
4463
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4464
+ "Params": 14,
4465
+ "Release": "V3",
4466
+ "Thinking": "Dense"
4467
+ },
4468
+ {
4469
+ "Model": "Hermes-4-14B",
4470
+ "Model Type": "General",
4471
+ "Benchmark": "VerilogEval S2R",
4472
+ "Task": "Power",
4473
+ "Result": 41.73,
4474
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4475
+ "Params": 14,
4476
+ "Release": "V3",
4477
+ "Thinking": "Dense"
4478
+ },
4479
+ {
4480
+ "Model": "Hermes-4-14B",
4481
+ "Model Type": "General",
4482
+ "Benchmark": "RTLLM",
4483
+ "Task": "Power",
4484
+ "Result": 50.59,
4485
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4486
+ "Params": 14,
4487
+ "Release": "V3",
4488
+ "Thinking": "Dense"
4489
+ },
4490
+ {
4491
+ "Model": "Hermes-4-14B",
4492
+ "Model Type": "General",
4493
+ "Benchmark": "VerilogEval S2R",
4494
+ "Task": "Performance",
4495
+ "Result": 40.33,
4496
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4497
+ "Params": 14,
4498
+ "Release": "V3",
4499
+ "Thinking": "Dense"
4500
+ },
4501
+ {
4502
+ "Model": "Hermes-4-14B",
4503
+ "Model Type": "General",
4504
+ "Benchmark": "RTLLM",
4505
+ "Task": "Performance",
4506
+ "Result": 45.81,
4507
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4508
+ "Params": 14,
4509
+ "Release": "V3",
4510
+ "Thinking": "Dense"
4511
+ },
4512
+ {
4513
+ "Model": "Hermes-4-14B",
4514
+ "Model Type": "General",
4515
+ "Benchmark": "VerilogEval S2R",
4516
+ "Task": "Area",
4517
+ "Result": 41.38,
4518
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4519
+ "Params": 14,
4520
+ "Release": "V3",
4521
+ "Thinking": "Dense"
4522
+ },
4523
+ {
4524
+ "Model": "Hermes-4-14B",
4525
+ "Model Type": "General",
4526
+ "Benchmark": "RTLLM",
4527
+ "Task": "Area",
4528
+ "Result": 47.91,
4529
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4530
+ "Params": 14,
4531
+ "Release": "V3",
4532
+ "Thinking": "Dense"
4533
+ },
4534
+ {
4535
+ "Model": "Hermes-4-14B",
4536
+ "Model Type": "General",
4537
+ "Benchmark": "RTL-Repo",
4538
+ "Task": "Exact Matching (EM)",
4539
+ "Result": 28.36,
4540
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4541
+ "Params": 14,
4542
+ "Release": "V3",
4543
+ "Thinking": "Dense"
4544
+ },
4545
+ {
4546
+ "Model": "Hermes-4-14B",
4547
+ "Model Type": "General",
4548
+ "Benchmark": "VerilogEval MC",
4549
+ "Task": "Syntax (STX)",
4550
+ "Result": 52.69,
4551
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4552
+ "Params": 14,
4553
+ "Release": "V3",
4554
+ "Thinking": "Dense"
4555
+ },
4556
+ {
4557
+ "Model": "Hermes-4-14B",
4558
+ "Model Type": "General",
4559
+ "Benchmark": "VeriGen",
4560
+ "Task": "Syntax (STX)",
4561
+ "Result": 83.53,
4562
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4563
+ "Params": 14,
4564
+ "Release": "V3",
4565
+ "Thinking": "Dense"
4566
+ },
4567
+ {
4568
+ "Model": "Hermes-4-14B",
4569
+ "Model Type": "General",
4570
+ "Benchmark": "VerilogEval MC",
4571
+ "Task": "Functionality (FNC)",
4572
+ "Result": 27.44,
4573
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4574
+ "Params": 14,
4575
+ "Release": "V3",
4576
+ "Thinking": "Dense"
4577
+ },
4578
+ {
4579
+ "Model": "Hermes-4-14B",
4580
+ "Model Type": "General",
4581
+ "Benchmark": "VeriGen",
4582
+ "Task": "Functionality (FNC)",
4583
+ "Result": 45.88,
4584
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4585
+ "Params": 14,
4586
+ "Release": "V3",
4587
+ "Thinking": "Dense"
4588
+ },
4589
+ {
4590
+ "Model": "Hermes-4-14B",
4591
+ "Model Type": "General",
4592
+ "Benchmark": "VerilogEval MC",
4593
+ "Task": "Synthesis (SYN)",
4594
+ "Result": 27.44,
4595
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4596
+ "Params": 14,
4597
+ "Release": "V3",
4598
+ "Thinking": "Dense"
4599
+ },
4600
+ {
4601
+ "Model": "Hermes-4-14B",
4602
+ "Model Type": "General",
4603
+ "Benchmark": "VeriGen",
4604
+ "Task": "Synthesis (SYN)",
4605
+ "Result": 45.88,
4606
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4607
+ "Params": 14,
4608
+ "Release": "V3",
4609
+ "Thinking": "Dense"
4610
+ },
4611
+ {
4612
+ "Model": "Hermes-4-14B",
4613
+ "Model Type": "General",
4614
+ "Benchmark": "VerilogEval MC",
4615
+ "Task": "Power",
4616
+ "Result": 26.54,
4617
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4618
+ "Params": 14,
4619
+ "Release": "V3",
4620
+ "Thinking": "Dense"
4621
+ },
4622
+ {
4623
+ "Model": "Hermes-4-14B",
4624
+ "Model Type": "General",
4625
+ "Benchmark": "VeriGen",
4626
+ "Task": "Power",
4627
+ "Result": 43.55,
4628
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4629
+ "Params": 14,
4630
+ "Release": "V3",
4631
+ "Thinking": "Dense"
4632
+ },
4633
+ {
4634
+ "Model": "Hermes-4-14B",
4635
+ "Model Type": "General",
4636
+ "Benchmark": "VerilogEval MC",
4637
+ "Task": "Performance",
4638
+ "Result": 26.68,
4639
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4640
+ "Params": 14,
4641
+ "Release": "V3",
4642
+ "Thinking": "Dense"
4643
+ },
4644
+ {
4645
+ "Model": "Hermes-4-14B",
4646
+ "Model Type": "General",
4647
+ "Benchmark": "VeriGen",
4648
+ "Task": "Performance",
4649
+ "Result": 45.98,
4650
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4651
+ "Params": 14,
4652
+ "Release": "V3",
4653
+ "Thinking": "Dense"
4654
+ },
4655
+ {
4656
+ "Model": "Hermes-4-14B",
4657
+ "Model Type": "General",
4658
+ "Benchmark": "VerilogEval MC",
4659
+ "Task": "Area",
4660
+ "Result": 25.11,
4661
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4662
+ "Params": 14,
4663
+ "Release": "V3",
4664
+ "Thinking": "Dense"
4665
+ },
4666
+ {
4667
+ "Model": "Hermes-4-14B",
4668
+ "Model Type": "General",
4669
+ "Benchmark": "VeriGen",
4670
+ "Task": "Area",
4671
+ "Result": 44.96,
4672
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4673
+ "Params": 14,
4674
+ "Release": "V3",
4675
+ "Thinking": "Dense"
4676
+ },
4677
+ {
4678
+ "Model": "Qwen3-8B",
4679
+ "Model Type": "General",
4680
+ "Benchmark": "VerilogEval S2R",
4681
+ "Task": "Syntax (STX)",
4682
+ "Result": 70.77,
4683
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4684
+ "Params": 8.2,
4685
+ "Release": "V3",
4686
+ "Thinking": "Reasoning"
4687
+ },
4688
+ {
4689
+ "Model": "Qwen3-8B",
4690
+ "Model Type": "General",
4691
+ "Benchmark": "RTLLM",
4692
+ "Task": "Syntax (STX)",
4693
+ "Result": 62.13,
4694
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4695
+ "Params": 8.2,
4696
+ "Release": "V3",
4697
+ "Thinking": "Reasoning"
4698
+ },
4699
+ {
4700
+ "Model": "Qwen3-8B",
4701
+ "Model Type": "General",
4702
+ "Benchmark": "VerilogEval S2R",
4703
+ "Task": "Functionality (FNC)",
4704
+ "Result": 50.0,
4705
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4706
+ "Params": 8.2,
4707
+ "Release": "V3",
4708
+ "Thinking": "Reasoning"
4709
+ },
4710
+ {
4711
+ "Model": "Qwen3-8B",
4712
+ "Model Type": "General",
4713
+ "Benchmark": "RTLLM",
4714
+ "Task": "Functionality (FNC)",
4715
+ "Result": 34.47,
4716
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4717
+ "Params": 8.2,
4718
+ "Release": "V3",
4719
+ "Thinking": "Reasoning"
4720
+ },
4721
+ {
4722
+ "Model": "Qwen3-8B",
4723
+ "Model Type": "General",
4724
+ "Benchmark": "VerilogEval S2R",
4725
+ "Task": "Synthesis (SYN)",
4726
+ "Result": 50.0,
4727
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4728
+ "Params": 8.2,
4729
+ "Release": "V3",
4730
+ "Thinking": "Reasoning"
4731
+ },
4732
+ {
4733
+ "Model": "Qwen3-8B",
4734
+ "Model Type": "General",
4735
+ "Benchmark": "RTLLM",
4736
+ "Task": "Synthesis (SYN)",
4737
+ "Result": 32.77,
4738
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4739
+ "Params": 8.2,
4740
+ "Release": "V3",
4741
+ "Thinking": "Reasoning"
4742
+ },
4743
+ {
4744
+ "Model": "Qwen3-8B",
4745
+ "Model Type": "General",
4746
+ "Benchmark": "VerilogEval S2R",
4747
+ "Task": "Power",
4748
+ "Result": 49.33,
4749
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4750
+ "Params": 8.2,
4751
+ "Release": "V3",
4752
+ "Thinking": "Reasoning"
4753
+ },
4754
+ {
4755
+ "Model": "Qwen3-8B",
4756
+ "Model Type": "General",
4757
+ "Benchmark": "RTLLM",
4758
+ "Task": "Power",
4759
+ "Result": 35.1,
4760
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4761
+ "Params": 8.2,
4762
+ "Release": "V3",
4763
+ "Thinking": "Reasoning"
4764
+ },
4765
+ {
4766
+ "Model": "Qwen3-8B",
4767
+ "Model Type": "General",
4768
+ "Benchmark": "VerilogEval S2R",
4769
+ "Task": "Performance",
4770
+ "Result": 47.92,
4771
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4772
+ "Params": 8.2,
4773
+ "Release": "V3",
4774
+ "Thinking": "Reasoning"
4775
+ },
4776
+ {
4777
+ "Model": "Qwen3-8B",
4778
+ "Model Type": "General",
4779
+ "Benchmark": "RTLLM",
4780
+ "Task": "Performance",
4781
+ "Result": 32.52,
4782
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4783
+ "Params": 8.2,
4784
+ "Release": "V3",
4785
+ "Thinking": "Reasoning"
4786
+ },
4787
+ {
4788
+ "Model": "Qwen3-8B",
4789
+ "Model Type": "General",
4790
+ "Benchmark": "VerilogEval S2R",
4791
+ "Task": "Area",
4792
+ "Result": 48.91,
4793
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4794
+ "Params": 8.2,
4795
+ "Release": "V3",
4796
+ "Thinking": "Reasoning"
4797
+ },
4798
+ {
4799
+ "Model": "Qwen3-8B",
4800
+ "Model Type": "General",
4801
+ "Benchmark": "RTLLM",
4802
+ "Task": "Area",
4803
+ "Result": 31.88,
4804
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4805
+ "Params": 8.2,
4806
+ "Release": "V3",
4807
+ "Thinking": "Reasoning"
4808
+ },
4809
+ {
4810
+ "Model": "Qwen3-8B",
4811
+ "Model Type": "General",
4812
+ "Benchmark": "RTL-Repo",
4813
+ "Task": "Exact Matching (EM)",
4814
+ "Result": -1.0,
4815
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4816
+ "Params": 8.2,
4817
+ "Release": "V3",
4818
+ "Thinking": "Reasoning"
4819
+ },
4820
+ {
4821
+ "Model": "Qwen3-8B",
4822
+ "Model Type": "General",
4823
+ "Benchmark": "VerilogEval MC",
4824
+ "Task": "Syntax (STX)",
4825
+ "Result": 69.36,
4826
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4827
+ "Params": 8.2,
4828
+ "Release": "V3",
4829
+ "Thinking": "Reasoning"
4830
+ },
4831
+ {
4832
+ "Model": "Qwen3-8B",
4833
+ "Model Type": "General",
4834
+ "Benchmark": "VeriGen",
4835
+ "Task": "Syntax (STX)",
4836
+ "Result": 81.18,
4837
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4838
+ "Params": 8.2,
4839
+ "Release": "V3",
4840
+ "Thinking": "Reasoning"
4841
+ },
4842
+ {
4843
+ "Model": "Qwen3-8B",
4844
+ "Model Type": "General",
4845
+ "Benchmark": "VerilogEval MC",
4846
+ "Task": "Functionality (FNC)",
4847
+ "Result": 51.54,
4848
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4849
+ "Params": 8.2,
4850
+ "Release": "V3",
4851
+ "Thinking": "Reasoning"
4852
+ },
4853
+ {
4854
+ "Model": "Qwen3-8B",
4855
+ "Model Type": "General",
4856
+ "Benchmark": "VeriGen",
4857
+ "Task": "Functionality (FNC)",
4858
+ "Result": 40.0,
4859
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4860
+ "Params": 8.2,
4861
+ "Release": "V3",
4862
+ "Thinking": "Reasoning"
4863
+ },
4864
+ {
4865
+ "Model": "Qwen3-8B",
4866
+ "Model Type": "General",
4867
+ "Benchmark": "VerilogEval MC",
4868
+ "Task": "Synthesis (SYN)",
4869
+ "Result": 50.64,
4870
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4871
+ "Params": 8.2,
4872
+ "Release": "V3",
4873
+ "Thinking": "Reasoning"
4874
+ },
4875
+ {
4876
+ "Model": "Qwen3-8B",
4877
+ "Model Type": "General",
4878
+ "Benchmark": "VeriGen",
4879
+ "Task": "Synthesis (SYN)",
4880
+ "Result": 38.82,
4881
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4882
+ "Params": 8.2,
4883
+ "Release": "V3",
4884
+ "Thinking": "Reasoning"
4885
+ },
4886
+ {
4887
+ "Model": "Qwen3-8B",
4888
+ "Model Type": "General",
4889
+ "Benchmark": "VerilogEval MC",
4890
+ "Task": "Power",
4891
+ "Result": 49.91,
4892
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4893
+ "Params": 8.2,
4894
+ "Release": "V3",
4895
+ "Thinking": "Reasoning"
4896
+ },
4897
+ {
4898
+ "Model": "Qwen3-8B",
4899
+ "Model Type": "General",
4900
+ "Benchmark": "VeriGen",
4901
+ "Task": "Power",
4902
+ "Result": 37.47,
4903
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4904
+ "Params": 8.2,
4905
+ "Release": "V3",
4906
+ "Thinking": "Reasoning"
4907
+ },
4908
+ {
4909
+ "Model": "Qwen3-8B",
4910
+ "Model Type": "General",
4911
+ "Benchmark": "VerilogEval MC",
4912
+ "Task": "Performance",
4913
+ "Result": 48.4,
4914
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4915
+ "Params": 8.2,
4916
+ "Release": "V3",
4917
+ "Thinking": "Reasoning"
4918
+ },
4919
+ {
4920
+ "Model": "Qwen3-8B",
4921
+ "Model Type": "General",
4922
+ "Benchmark": "VeriGen",
4923
+ "Task": "Performance",
4924
+ "Result": 38.76,
4925
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4926
+ "Params": 8.2,
4927
+ "Release": "V3",
4928
+ "Thinking": "Reasoning"
4929
+ },
4930
+ {
4931
+ "Model": "Qwen3-8B",
4932
+ "Model Type": "General",
4933
+ "Benchmark": "VerilogEval MC",
4934
+ "Task": "Area",
4935
+ "Result": 49.41,
4936
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4937
+ "Params": 8.2,
4938
+ "Release": "V3",
4939
+ "Thinking": "Reasoning"
4940
+ },
4941
+ {
4942
+ "Model": "Qwen3-8B",
4943
+ "Model Type": "General",
4944
+ "Benchmark": "VeriGen",
4945
+ "Task": "Area",
4946
+ "Result": 38.73,
4947
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4948
+ "Params": 8.2,
4949
+ "Release": "V3",
4950
+ "Thinking": "Reasoning"
4951
+ },
4952
  {
4953
  "Model": "CodeLlama 70B",
4954
  "Model Type": "Coding",
results/results_icarus_october_2025.csv CHANGED
@@ -33,8 +33,8 @@ RTLCoder DeepSeek,84.620000000000005,73.060000000000002,39.490000000000002,37.14
33
  OriGen,96.150000000000006,81.629999999999995,54.229999999999997,50.609999999999999,54.229999999999997,50.609999999999999,54.289999999999999,53.100000000000001,51.57,50.859999999999999,53.149999999999999,53.439999999999998,17.07,92.439999999999998,98.819999999999993,50.770000000000003,76.469999999999999,50.770000000000003,76.469999999999999,50.950000000000003,71.659999999999997,48.530000000000001,76.609999999999999,50.509999999999998,78.700000000000003
34
  CodeV-CL-7B,32.18,48.159999999999997,13.08,24.489999999999998,12.949999999999999,21.629999999999999,12.800000000000001,22.25,12.51,20.59,12.82,21.289999999999999,12.27,92.049999999999997,98.819999999999993,31.789999999999999,49.409999999999997,31.789999999999999,49.409999999999997,31.739999999999998,48.130000000000003,29.449999999999999,49.340000000000003,31.609999999999999,49.079999999999998
35
  CodeV-DS-6.7B,33.590000000000003,67.349999999999994,15,38.780000000000001,15,37.140000000000001,15.1,35.560000000000002,14.460000000000001,35.130000000000003,14.85,35.880000000000003,21.260000000000002,95.510000000000005,100,47.049999999999997,61.18,47.049999999999997,60,47.369999999999997,59.880000000000003,44.350000000000001,59.960000000000001,46.520000000000003,59.770000000000003
36
- ,,,,,,,,,,,,,,,,,,,,,,,,,
37
- ,,,,,,,,,,,,,,,,,,,,,,,,,
38
  ,,,,,,,,,,,,,,,,,,,,,,,,,
39
  ,,,,,,,,,,,,,,,,,,,,,,,,,
40
  ,,,,,,,,,,,,,,,,,,,,,,,,,
 
33
  OriGen,96.150000000000006,81.629999999999995,54.229999999999997,50.609999999999999,54.229999999999997,50.609999999999999,54.289999999999999,53.100000000000001,51.57,50.859999999999999,53.149999999999999,53.439999999999998,17.07,92.439999999999998,98.819999999999993,50.770000000000003,76.469999999999999,50.770000000000003,76.469999999999999,50.950000000000003,71.659999999999997,48.530000000000001,76.609999999999999,50.509999999999998,78.700000000000003
34
  CodeV-CL-7B,32.18,48.159999999999997,13.08,24.489999999999998,12.949999999999999,21.629999999999999,12.800000000000001,22.25,12.51,20.59,12.82,21.289999999999999,12.27,92.049999999999997,98.819999999999993,31.789999999999999,49.409999999999997,31.789999999999999,49.409999999999997,31.739999999999998,48.130000000000003,29.449999999999999,49.340000000000003,31.609999999999999,49.079999999999998
35
  CodeV-DS-6.7B,33.590000000000003,67.349999999999994,15,38.780000000000001,15,37.140000000000001,15.1,35.560000000000002,14.460000000000001,35.130000000000003,14.85,35.880000000000003,21.260000000000002,95.510000000000005,100,47.049999999999997,61.18,47.049999999999997,60,47.369999999999997,59.880000000000003,44.350000000000001,59.960000000000001,46.520000000000003,59.770000000000003
36
+ Qwen3-8B,70.769999999999996,62.130000000000003,50,34.469999999999999,50,32.770000000000003,49.329999999999998,35.100000000000001,47.920000000000002,32.520000000000003,48.909999999999997,31.879999999999999,-1,69.359999999999999,81.180000000000007,51.539999999999999,40,50.640000000000001,38.82,49.909999999999997,37.469999999999999,48.399999999999999,38.759999999999998,49.409999999999997,38.729999999999997
37
+ Hermes-4-14B,72.180000000000007,80.849999999999994,42.310000000000002,48.090000000000003,42.310000000000002,45.530000000000001,41.729999999999997,50.590000000000003,40.329999999999998,45.810000000000002,41.380000000000003,47.909999999999997,-1,52.689999999999998,83.530000000000001,27.440000000000001,45.880000000000003,27.440000000000001,45.880000000000003,26.539999999999999,43.549999999999997,26.68,45.979999999999997,25.109999999999999,44.960000000000001
38
  ,,,,,,,,,,,,,,,,,,,,,,,,,
39
  ,,,,,,,,,,,,,,,,,,,,,,,,,
40
  ,,,,,,,,,,,,,,,,,,,,,,,,,
results/results_v3_mlcad_icarus.csv CHANGED
@@ -8,12 +8,16 @@ Qwen3 236B A22B,91.280000000000001,73.879999999999995,76.920000000000002,51.43,7
8
  gpt-oss-120b,92.950000000000003,83.829999999999998,78.079999999999998,60,77.560000000000002,51.490000000000002,76.659999999999997,53.960000000000001,74.489999999999995,49.850000000000001,76.530000000000001,54.640000000000001,-1,95.640000000000001,88.239999999999995,81.150000000000006,68.239999999999995,80.510000000000005,68.239999999999995,79.719999999999999,70.099999999999994,77.540000000000006,68.269999999999996,78.969999999999999,69.849999999999994
9
  Qwen2.5 72B,82.180000000000007,79.590000000000003,52.439999999999998,45.310000000000002,51.920000000000002,44.079999999999998,51.829999999999998,46.469999999999999,48.75,45.399999999999999,50.090000000000003,47.649999999999999,37.439999999999998,80.900000000000006,84.709999999999994,52.950000000000003,47.060000000000002,52.689999999999998,47.060000000000002,51.659999999999997,47.590000000000003,49.369999999999997,46.960000000000001,51.18,47.700000000000003
10
  Llama 3.(1-3) 70B,66.150000000000006,73.879999999999995,40.640000000000001,42.450000000000003,40.640000000000001,39.18,40.460000000000001,40.810000000000002,38.079999999999998,38.140000000000001,39.859999999999999,39.649999999999999,28.719999999999999,84.739999999999995,89.409999999999997,41.670000000000002,65.879999999999995,41.670000000000002,64.709999999999994,41.380000000000003,63.469999999999999,39.75,64.689999999999998,41.359999999999999,64.709999999999994
 
11
  QwQ 32B,87.950000000000003,82.450000000000003,66.409999999999997,56.729999999999997,66.409999999999997,52.240000000000002,66.150000000000006,55.829999999999998,63.799999999999997,51.909999999999997,65.120000000000005,56.07,-1,58.969999999999999,68.239999999999995,40,52.939999999999998,39.619999999999997,52.939999999999998,39.399999999999999,51.469999999999999,37.530000000000001,52.93,39.100000000000001,53.390000000000001
12
  Qwen2.5 32B,88.590000000000003,84.079999999999998,52.560000000000002,50.200000000000003,52.18,46.119999999999997,52.32,49.729999999999997,49.43,46.43,50.82,50.43,28.93,93.209999999999994,85.879999999999995,41.539999999999999,45.880000000000003,41.539999999999999,45.880000000000003,41.310000000000002,43.560000000000002,40.479999999999997,46.079999999999998,41.229999999999997,45.369999999999997
13
  Magistral Small 2506,63.850000000000001,71.909999999999997,42.950000000000003,43.399999999999999,42.439999999999998,37.450000000000003,41.390000000000001,43.359999999999999,40.090000000000003,37.439999999999998,41.390000000000001,40.32,-1,31.280000000000001,68.239999999999995,20.899999999999999,50.590000000000003,20.77,49.409999999999997,20.109999999999999,41.590000000000003,19.609999999999999,48.969999999999999,20.16,50.07
14
  gpt-oss-20b,88.079999999999998,86.810000000000002,69.739999999999995,58.719999999999999,69.230000000000004,48.939999999999998,69.849999999999994,55.890000000000001,64.569999999999993,45.210000000000001,67.819999999999993,53.289999999999999,-1,91.920000000000002,76.469999999999999,70.260000000000005,50.590000000000003,69.739999999999995,50.590000000000003,69.5,45.409999999999997,67.060000000000002,50.479999999999997,68.760000000000005,49.850000000000001
15
  StarChat2 15B v0.1,88.459999999999994,84.900000000000006,37.950000000000003,44.490000000000002,37.950000000000003,44.079999999999998,37.560000000000002,46.950000000000003,35.299999999999997,43.219999999999999,37.189999999999998,46.649999999999999,13.42,79.739999999999995,92.939999999999998,36.409999999999997,63.530000000000001,36.030000000000001,63.530000000000001,36.079999999999998,58.060000000000002,34.909999999999997,63.259999999999998,35.759999999999998,64.560000000000002
16
  DeepSeek R1 Distill Qwen 14B,42.18,34.689999999999998,25.510000000000002,18.370000000000001,25.510000000000002,16.329999999999998,25.359999999999999,17.859999999999999,24.190000000000001,16.48,25.27,17.329999999999998,-1,45,44.710000000000001,25.640000000000001,28.239999999999998,25.260000000000002,28.239999999999998,24.789999999999999,24.710000000000001,23.48,28.140000000000001,24.629999999999999,28.350000000000001
 
 
 
17
  CodeLlama 70B,67.049999999999997,69.799999999999997,33.079999999999998,36.329999999999998,33.079999999999998,34.289999999999999,32.689999999999998,37.189999999999998,31.460000000000001,34.289999999999999,32.439999999999998,35.950000000000003,24.329999999999998,90.769999999999996,88.239999999999995,33.329999999999998,47.060000000000002,33.329999999999998,47.060000000000002,33.020000000000003,45.799999999999997,30.800000000000001,46.909999999999997,32.990000000000002,46.979999999999997
18
  DeepSeek Coder 33B,62.82,83.670000000000002,23.329999999999998,42.450000000000003,23.079999999999998,42.039999999999999,22.859999999999999,42.289999999999999,22.809999999999999,39.420000000000002,22.289999999999999,42.710000000000001,24.579999999999998,75.260000000000005,88.239999999999995,39.619999999999997,45.880000000000003,39.359999999999999,45.880000000000003,38.229999999999997,46.259999999999998,36.789999999999999,45.609999999999999,37.899999999999999,46.170000000000002
19
  QwenCoder 2.5 32B,87.180000000000007,77.959999999999994,45,43.270000000000003,44.869999999999997,43.270000000000003,44.25,46.82,43.030000000000001,43.200000000000003,43.759999999999998,45.420000000000002,31.07,83.719999999999999,87.060000000000002,45.640000000000001,54.119999999999997,45.130000000000003,54.119999999999997,44.590000000000003,54.549999999999997,43.009999999999998,54.009999999999998,44.549999999999997,55.009999999999998
@@ -61,6 +65,4 @@ CodeV-DS-6.7B,33.590000000000003,67.349999999999994,15,38.780000000000001,15,37.
61
  ,,,,,,,,,,,,,,,,,,,,,,,,,
62
  ,,,,,,,,,,,,,,,,,,,,,,,,,
63
  ,,,,,,,,,,,,,,,,,,,,,,,,,
64
- ,,,,,,,,,,,,,,,,,,,,,,,,,
65
- ,,,,,,,,,,,,,,,,,,,,,,,,,
66
 
 
8
  gpt-oss-120b,92.950000000000003,83.829999999999998,78.079999999999998,60,77.560000000000002,51.490000000000002,76.659999999999997,53.960000000000001,74.489999999999995,49.850000000000001,76.530000000000001,54.640000000000001,-1,95.640000000000001,88.239999999999995,81.150000000000006,68.239999999999995,80.510000000000005,68.239999999999995,79.719999999999999,70.099999999999994,77.540000000000006,68.269999999999996,78.969999999999999,69.849999999999994
9
  Qwen2.5 72B,82.180000000000007,79.590000000000003,52.439999999999998,45.310000000000002,51.920000000000002,44.079999999999998,51.829999999999998,46.469999999999999,48.75,45.399999999999999,50.090000000000003,47.649999999999999,37.439999999999998,80.900000000000006,84.709999999999994,52.950000000000003,47.060000000000002,52.689999999999998,47.060000000000002,51.659999999999997,47.590000000000003,49.369999999999997,46.960000000000001,51.18,47.700000000000003
10
  Llama 3.(1-3) 70B,66.150000000000006,73.879999999999995,40.640000000000001,42.450000000000003,40.640000000000001,39.18,40.460000000000001,40.810000000000002,38.079999999999998,38.140000000000001,39.859999999999999,39.649999999999999,28.719999999999999,84.739999999999995,89.409999999999997,41.670000000000002,65.879999999999995,41.670000000000002,64.709999999999994,41.380000000000003,63.469999999999999,39.75,64.689999999999998,41.359999999999999,64.709999999999994
11
+ Seed-OSS-36B,88.969999999999999,81.700000000000003,71.150000000000006,59.149999999999999,71.150000000000006,56.600000000000001,70.099999999999994,64.629999999999995,68.099999999999994,57.32,70.319999999999993,60.880000000000003,-1,91.030000000000001,91.760000000000005,76.540000000000006,70.590000000000003,76.150000000000006,70.590000000000003,74.950000000000003,66.109999999999999,73.329999999999998,70.189999999999998,74.689999999999998,72.450000000000003
12
  QwQ 32B,87.950000000000003,82.450000000000003,66.409999999999997,56.729999999999997,66.409999999999997,52.240000000000002,66.150000000000006,55.829999999999998,63.799999999999997,51.909999999999997,65.120000000000005,56.07,-1,58.969999999999999,68.239999999999995,40,52.939999999999998,39.619999999999997,52.939999999999998,39.399999999999999,51.469999999999999,37.530000000000001,52.93,39.100000000000001,53.390000000000001
13
  Qwen2.5 32B,88.590000000000003,84.079999999999998,52.560000000000002,50.200000000000003,52.18,46.119999999999997,52.32,49.729999999999997,49.43,46.43,50.82,50.43,28.93,93.209999999999994,85.879999999999995,41.539999999999999,45.880000000000003,41.539999999999999,45.880000000000003,41.310000000000002,43.560000000000002,40.479999999999997,46.079999999999998,41.229999999999997,45.369999999999997
14
  Magistral Small 2506,63.850000000000001,71.909999999999997,42.950000000000003,43.399999999999999,42.439999999999998,37.450000000000003,41.390000000000001,43.359999999999999,40.090000000000003,37.439999999999998,41.390000000000001,40.32,-1,31.280000000000001,68.239999999999995,20.899999999999999,50.590000000000003,20.77,49.409999999999997,20.109999999999999,41.590000000000003,19.609999999999999,48.969999999999999,20.16,50.07
15
  gpt-oss-20b,88.079999999999998,86.810000000000002,69.739999999999995,58.719999999999999,69.230000000000004,48.939999999999998,69.849999999999994,55.890000000000001,64.569999999999993,45.210000000000001,67.819999999999993,53.289999999999999,-1,91.920000000000002,76.469999999999999,70.260000000000005,50.590000000000003,69.739999999999995,50.590000000000003,69.5,45.409999999999997,67.060000000000002,50.479999999999997,68.760000000000005,49.850000000000001
16
  StarChat2 15B v0.1,88.459999999999994,84.900000000000006,37.950000000000003,44.490000000000002,37.950000000000003,44.079999999999998,37.560000000000002,46.950000000000003,35.299999999999997,43.219999999999999,37.189999999999998,46.649999999999999,13.42,79.739999999999995,92.939999999999998,36.409999999999997,63.530000000000001,36.030000000000001,63.530000000000001,36.079999999999998,58.060000000000002,34.909999999999997,63.259999999999998,35.759999999999998,64.560000000000002
17
  DeepSeek R1 Distill Qwen 14B,42.18,34.689999999999998,25.510000000000002,18.370000000000001,25.510000000000002,16.329999999999998,25.359999999999999,17.859999999999999,24.190000000000001,16.48,25.27,17.329999999999998,-1,45,44.710000000000001,25.640000000000001,28.239999999999998,25.260000000000002,28.239999999999998,24.789999999999999,24.710000000000001,23.48,28.140000000000001,24.629999999999999,28.350000000000001
18
+ Hermes-4-14B-Reasoning,79.739999999999995,76.599999999999994,55.899999999999999,39.57,55.640000000000001,34.469999999999999,55.340000000000003,38.350000000000001,53.729999999999997,32.729999999999997,55.170000000000002,36.130000000000003,-1,51.149999999999999,74.120000000000005,36.539999999999999,56.469999999999999,36.280000000000001,56.469999999999999,35.259999999999998,52.969999999999999,35.520000000000003,56.219999999999999,34.939999999999998,57.439999999999998
19
+ Hermes-4-14B,72.180000000000007,80.849999999999994,42.310000000000002,48.090000000000003,42.310000000000002,45.530000000000001,41.729999999999997,50.590000000000003,40.329999999999998,45.810000000000002,41.380000000000003,47.909999999999997,28.359999999999999,52.689999999999998,83.530000000000001,27.440000000000001,45.880000000000003,27.440000000000001,45.880000000000003,26.539999999999999,43.549999999999997,26.68,45.979999999999997,25.109999999999999,44.960000000000001
20
+ Qwen3-8B,70.769999999999996,62.130000000000003,50,34.469999999999999,50,32.770000000000003,49.329999999999998,35.100000000000001,47.920000000000002,32.520000000000003,48.909999999999997,31.879999999999999,-1,69.359999999999999,81.180000000000007,51.539999999999999,40,50.640000000000001,38.82,49.909999999999997,37.469999999999999,48.399999999999999,38.759999999999998,49.409999999999997,38.729999999999997
21
  CodeLlama 70B,67.049999999999997,69.799999999999997,33.079999999999998,36.329999999999998,33.079999999999998,34.289999999999999,32.689999999999998,37.189999999999998,31.460000000000001,34.289999999999999,32.439999999999998,35.950000000000003,24.329999999999998,90.769999999999996,88.239999999999995,33.329999999999998,47.060000000000002,33.329999999999998,47.060000000000002,33.020000000000003,45.799999999999997,30.800000000000001,46.909999999999997,32.990000000000002,46.979999999999997
22
  DeepSeek Coder 33B,62.82,83.670000000000002,23.329999999999998,42.450000000000003,23.079999999999998,42.039999999999999,22.859999999999999,42.289999999999999,22.809999999999999,39.420000000000002,22.289999999999999,42.710000000000001,24.579999999999998,75.260000000000005,88.239999999999995,39.619999999999997,45.880000000000003,39.359999999999999,45.880000000000003,38.229999999999997,46.259999999999998,36.789999999999999,45.609999999999999,37.899999999999999,46.170000000000002
23
  QwenCoder 2.5 32B,87.180000000000007,77.959999999999994,45,43.270000000000003,44.869999999999997,43.270000000000003,44.25,46.82,43.030000000000001,43.200000000000003,43.759999999999998,45.420000000000002,31.07,83.719999999999999,87.060000000000002,45.640000000000001,54.119999999999997,45.130000000000003,54.119999999999997,44.590000000000003,54.549999999999997,43.009999999999998,54.009999999999998,44.549999999999997,55.009999999999998
 
65
  ,,,,,,,,,,,,,,,,,,,,,,,,,
66
  ,,,,,,,,,,,,,,,,,,,,,,,,,
67
  ,,,,,,,,,,,,,,,,,,,,,,,,,
 
 
68
 
results/results_v3_mlcad_verilator.csv CHANGED
@@ -8,12 +8,16 @@ Qwen3 236B A22B,93.969999999999999,78.719999999999999,75.769999999999996,56.1700
8
  gpt-oss-120b,94.359999999999999,85.959999999999994,76.670000000000002,62.130000000000003,76.540000000000006,51.490000000000002,76.510000000000005,53.960000000000001,73.829999999999998,49.850000000000001,76.010000000000005,54.640000000000001,-1,95.379999999999995,89.409999999999997,76.920000000000002,69.409999999999997,76.670000000000002,69.409999999999997,76.480000000000004,71.049999999999997,73.909999999999997,69.439999999999998,75.810000000000002,70.879999999999995
9
  Qwen2.5 72B,83.849999999999994,82.980000000000004,55,48.509999999999998,54.490000000000002,45.960000000000001,53.969999999999999,48.450000000000003,51.299999999999997,47.340000000000003,52.840000000000003,49.68,37.439999999999998,83.079999999999998,87.060000000000002,54.740000000000002,50.590000000000003,54.359999999999999,50.590000000000003,53.359999999999999,50.049999999999997,51.200000000000003,50.560000000000002,52.759999999999998,52.229999999999997
10
  Llama 3.(1-3) 70B,68.329999999999998,79.150000000000006,40.899999999999999,45.960000000000001,40.899999999999999,40.43,40.770000000000003,42.119999999999997,38.329999999999998,39.340000000000003,40.420000000000002,40.920000000000002,28.719999999999999,86.030000000000001,87.060000000000002,43.719999999999999,63.530000000000001,43.590000000000003,63.530000000000001,43.549999999999997,62.289999999999999,41.57,63.520000000000003,42.939999999999998,63.539999999999999
 
11
  QwQ 32B,89.230000000000004,85.959999999999994,67.560000000000002,58.299999999999997,67.180000000000007,53.619999999999997,67.040000000000006,57.289999999999999,64.780000000000001,53.340000000000003,66.439999999999998,57.560000000000002,-1,59.619999999999997,87.060000000000002,40.130000000000003,65.879999999999995,39.490000000000002,64.709999999999994,39.259999999999998,60.950000000000003,37.490000000000002,64.480000000000004,39.170000000000002,66.310000000000002
12
  Qwen2.5 32B,90.900000000000006,87.659999999999997,55.899999999999999,50.210000000000001,55.259999999999998,46.810000000000002,55.560000000000002,50.659999999999997,53.100000000000001,47.409999999999997,54.020000000000003,51.369999999999997,28.93,95.510000000000005,85.879999999999995,45,49.409999999999997,43.719999999999999,49.409999999999997,43.719999999999999,46.409999999999997,42.490000000000002,49.890000000000001,42.950000000000003,49.609999999999999
13
  Magistral Small 2506,65.129999999999995,74.040000000000006,42.950000000000003,44.259999999999998,42.310000000000002,37.450000000000003,41.810000000000002,43.100000000000001,40.189999999999998,37.369999999999997,41.75,40.350000000000001,-1,31.789999999999999,70.590000000000003,21.030000000000001,55.289999999999999,20.77,54.119999999999997,20.329999999999998,46.600000000000001,20.079999999999998,54.020000000000003,20.66,55.060000000000002
14
  gpt-oss-20b,88.209999999999994,89.790000000000006,68.079999999999998,61.280000000000001,67.689999999999998,49.359999999999999,68.920000000000002,57.140000000000001,63.630000000000003,45.880000000000003,66.810000000000002,54.340000000000003,-1,91.670000000000002,76.469999999999999,68.849999999999994,51.759999999999998,68.590000000000003,51.759999999999998,68.689999999999998,46.57,66.390000000000001,51.560000000000002,68.040000000000006,50.869999999999997
15
  StarChat2 15B v0.1,88.719999999999999,89.359999999999999,38.850000000000001,48.509999999999998,38.850000000000001,45.960000000000001,38.920000000000002,48.950000000000003,36.18,45.049999999999997,38.810000000000002,48.630000000000003,13.42,81.409999999999997,91.760000000000005,36.539999999999999,63.530000000000001,36.280000000000001,63.530000000000001,36.189999999999998,57.509999999999998,35.109999999999999,63.229999999999997,36.240000000000002,65
16
  DeepSeek R1 Distill Qwen 14B,42.82,36.600000000000001,25.260000000000002,20.43,25,17.02,24.809999999999999,18.620000000000001,23.68,17.18,24.829999999999998,18.07,-1,45.640000000000001,44.710000000000001,25.379999999999999,29.41,25.260000000000002,29.41,24.98,25.920000000000002,23.609999999999999,29.43,24.690000000000001,29.649999999999999
 
 
 
17
  CodeLlama 70B,67.310000000000002,77.450000000000003,34.740000000000002,38.719999999999999,34.490000000000002,36.170000000000002,34.030000000000001,39.18,32.880000000000003,36.140000000000001,33.829999999999998,37.93,24.329999999999998,92.560000000000002,88.239999999999995,35.770000000000003,52.939999999999998,35.380000000000003,51.759999999999998,34.810000000000002,50.619999999999997,32.340000000000003,51.890000000000001,34.520000000000003,52.359999999999999
18
  DeepSeek Coder 33B,64.489999999999995,88.939999999999998,23.59,46.380000000000003,23.59,44.259999999999998,23.370000000000001,44.539999999999999,23.32,41.539999999999999,22.739999999999998,45.039999999999999,24.579999999999998,76.540000000000006,88.239999999999995,40.130000000000003,50.590000000000003,39.869999999999997,50.590000000000003,38.700000000000003,51.5,37.329999999999998,50.219999999999999,38.399999999999999,51.630000000000003
19
  QwenCoder 2.5 32B,90.260000000000005,81.280000000000001,46.539999999999999,47.659999999999997,46.409999999999997,45.109999999999999,45.950000000000003,48.810000000000002,44.420000000000002,45.039999999999999,45.57,47.359999999999999,31.07,86.150000000000006,85.879999999999995,48.210000000000001,54.119999999999997,47.560000000000002,54.119999999999997,47.350000000000001,54.590000000000003,45.350000000000001,54.119999999999997,46.780000000000001,55.140000000000001
@@ -61,6 +65,4 @@ CodeV-DS-6.7B,34.359999999999999,69.790000000000006,14.74,41.280000000000001,14.
61
  ,,,,,,,,,,,,,,,,,,,,,,,,,
62
  ,,,,,,,,,,,,,,,,,,,,,,,,,
63
  ,,,,,,,,,,,,,,,,,,,,,,,,,
64
- ,,,,,,,,,,,,,,,,,,,,,,,,,
65
- ,,,,,,,,,,,,,,,,,,,,,,,,,
66
 
 
8
  gpt-oss-120b,94.359999999999999,85.959999999999994,76.670000000000002,62.130000000000003,76.540000000000006,51.490000000000002,76.510000000000005,53.960000000000001,73.829999999999998,49.850000000000001,76.010000000000005,54.640000000000001,-1,95.379999999999995,89.409999999999997,76.920000000000002,69.409999999999997,76.670000000000002,69.409999999999997,76.480000000000004,71.049999999999997,73.909999999999997,69.439999999999998,75.810000000000002,70.879999999999995
9
  Qwen2.5 72B,83.849999999999994,82.980000000000004,55,48.509999999999998,54.490000000000002,45.960000000000001,53.969999999999999,48.450000000000003,51.299999999999997,47.340000000000003,52.840000000000003,49.68,37.439999999999998,83.079999999999998,87.060000000000002,54.740000000000002,50.590000000000003,54.359999999999999,50.590000000000003,53.359999999999999,50.049999999999997,51.200000000000003,50.560000000000002,52.759999999999998,52.229999999999997
10
  Llama 3.(1-3) 70B,68.329999999999998,79.150000000000006,40.899999999999999,45.960000000000001,40.899999999999999,40.43,40.770000000000003,42.119999999999997,38.329999999999998,39.340000000000003,40.420000000000002,40.920000000000002,28.719999999999999,86.030000000000001,87.060000000000002,43.719999999999999,63.530000000000001,43.590000000000003,63.530000000000001,43.549999999999997,62.289999999999999,41.57,63.520000000000003,42.939999999999998,63.539999999999999
11
+ Seed-OSS-36B,89.359999999999999,83.400000000000006,71.150000000000006,61.700000000000003,70.900000000000006,57.020000000000003,70.530000000000001,65.329999999999998,68.219999999999999,57.32,70.329999999999998,61.390000000000001,-1,91.030000000000001,91.760000000000005,73.849999999999994,70.590000000000003,73.459999999999994,70.590000000000003,73.040000000000006,66.109999999999999,70.890000000000001,70.189999999999998,72.599999999999994,72.450000000000003
12
  QwQ 32B,89.230000000000004,85.959999999999994,67.560000000000002,58.299999999999997,67.180000000000007,53.619999999999997,67.040000000000006,57.289999999999999,64.780000000000001,53.340000000000003,66.439999999999998,57.560000000000002,-1,59.619999999999997,87.060000000000002,40.130000000000003,65.879999999999995,39.490000000000002,64.709999999999994,39.259999999999998,60.950000000000003,37.490000000000002,64.480000000000004,39.170000000000002,66.310000000000002
13
  Qwen2.5 32B,90.900000000000006,87.659999999999997,55.899999999999999,50.210000000000001,55.259999999999998,46.810000000000002,55.560000000000002,50.659999999999997,53.100000000000001,47.409999999999997,54.020000000000003,51.369999999999997,28.93,95.510000000000005,85.879999999999995,45,49.409999999999997,43.719999999999999,49.409999999999997,43.719999999999999,46.409999999999997,42.490000000000002,49.890000000000001,42.950000000000003,49.609999999999999
14
  Magistral Small 2506,65.129999999999995,74.040000000000006,42.950000000000003,44.259999999999998,42.310000000000002,37.450000000000003,41.810000000000002,43.100000000000001,40.189999999999998,37.369999999999997,41.75,40.350000000000001,-1,31.789999999999999,70.590000000000003,21.030000000000001,55.289999999999999,20.77,54.119999999999997,20.329999999999998,46.600000000000001,20.079999999999998,54.020000000000003,20.66,55.060000000000002
15
  gpt-oss-20b,88.209999999999994,89.790000000000006,68.079999999999998,61.280000000000001,67.689999999999998,49.359999999999999,68.920000000000002,57.140000000000001,63.630000000000003,45.880000000000003,66.810000000000002,54.340000000000003,-1,91.670000000000002,76.469999999999999,68.849999999999994,51.759999999999998,68.590000000000003,51.759999999999998,68.689999999999998,46.57,66.390000000000001,51.560000000000002,68.040000000000006,50.869999999999997
16
  StarChat2 15B v0.1,88.719999999999999,89.359999999999999,38.850000000000001,48.509999999999998,38.850000000000001,45.960000000000001,38.920000000000002,48.950000000000003,36.18,45.049999999999997,38.810000000000002,48.630000000000003,13.42,81.409999999999997,91.760000000000005,36.539999999999999,63.530000000000001,36.280000000000001,63.530000000000001,36.189999999999998,57.509999999999998,35.109999999999999,63.229999999999997,36.240000000000002,65
17
  DeepSeek R1 Distill Qwen 14B,42.82,36.600000000000001,25.260000000000002,20.43,25,17.02,24.809999999999999,18.620000000000001,23.68,17.18,24.829999999999998,18.07,-1,45.640000000000001,44.710000000000001,25.379999999999999,29.41,25.260000000000002,29.41,24.98,25.920000000000002,23.609999999999999,29.43,24.690000000000001,29.649999999999999
18
+ Hermes-4-14B-Reasoning,80.640000000000001,79.569999999999993,56.149999999999999,41.280000000000001,55.770000000000003,34.469999999999999,55.850000000000001,38.350000000000001,53.710000000000001,32.729999999999997,55.649999999999999,36.130000000000003,-1,52.18,84.709999999999994,36.920000000000002,60,36.670000000000002,58.82,36.210000000000001,52.969999999999999,36.020000000000003,58.57,36.119999999999997,59.799999999999997
19
+ Hermes-4-14B,72.180000000000007,85.109999999999999,43.850000000000001,50.640000000000001,43.850000000000001,45.960000000000001,43.270000000000003,51.009999999999998,41.869999999999997,46.229999999999997,42.920000000000002,48.329999999999998,28.359999999999999,54.229999999999997,84.709999999999994,27.690000000000001,48.240000000000002,27.690000000000001,48.240000000000002,27.359999999999999,46.18,27.34,48.57,26.199999999999999,47.469999999999999
20
+ Qwen3-8B,70.900000000000006,65.959999999999994,51.030000000000001,37.869999999999997,51.030000000000001,33.619999999999997,50.57,36.520000000000003,48.899999999999999,33.600000000000001,49.93,33.200000000000003,-1,70.769999999999996,83.530000000000001,51.409999999999997,40,51.149999999999999,38.82,50.579999999999998,37.469999999999999,49.170000000000002,38.759999999999998,50.18,38.729999999999997
21
  CodeLlama 70B,67.310000000000002,77.450000000000003,34.740000000000002,38.719999999999999,34.490000000000002,36.170000000000002,34.030000000000001,39.18,32.880000000000003,36.140000000000001,33.829999999999998,37.93,24.329999999999998,92.560000000000002,88.239999999999995,35.770000000000003,52.939999999999998,35.380000000000003,51.759999999999998,34.810000000000002,50.619999999999997,32.340000000000003,51.890000000000001,34.520000000000003,52.359999999999999
22
  DeepSeek Coder 33B,64.489999999999995,88.939999999999998,23.59,46.380000000000003,23.59,44.259999999999998,23.370000000000001,44.539999999999999,23.32,41.539999999999999,22.739999999999998,45.039999999999999,24.579999999999998,76.540000000000006,88.239999999999995,40.130000000000003,50.590000000000003,39.869999999999997,50.590000000000003,38.700000000000003,51.5,37.329999999999998,50.219999999999999,38.399999999999999,51.630000000000003
23
  QwenCoder 2.5 32B,90.260000000000005,81.280000000000001,46.539999999999999,47.659999999999997,46.409999999999997,45.109999999999999,45.950000000000003,48.810000000000002,44.420000000000002,45.039999999999999,45.57,47.359999999999999,31.07,86.150000000000006,85.879999999999995,48.210000000000001,54.119999999999997,47.560000000000002,54.119999999999997,47.350000000000001,54.590000000000003,45.350000000000001,54.119999999999997,46.780000000000001,55.140000000000001
 
65
  ,,,,,,,,,,,,,,,,,,,,,,,,,
66
  ,,,,,,,,,,,,,,,,,,,,,,,,,
67
  ,,,,,,,,,,,,,,,,,,,,,,,,,
 
 
68
 
results/results_verilator.json CHANGED
@@ -4124,6 +4124,831 @@
4124
  "Release": "V1",
4125
  "Thinking": "Reasoning"
4126
  },
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4127
  {
4128
  "Model": "CodeLlama 70B",
4129
  "Model Type": "Coding",
 
4124
  "Release": "V1",
4125
  "Thinking": "Reasoning"
4126
  },
4127
+ {
4128
+ "Model": "Hermes-4-14B-Reasoning",
4129
+ "Model Type": "General",
4130
+ "Benchmark": "VerilogEval S2R",
4131
+ "Task": "Syntax (STX)",
4132
+ "Result": 80.64,
4133
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4134
+ "Params": 14,
4135
+ "Release": "V3",
4136
+ "Thinking": "Reasoning"
4137
+ },
4138
+ {
4139
+ "Model": "Hermes-4-14B-Reasoning",
4140
+ "Model Type": "General",
4141
+ "Benchmark": "RTLLM",
4142
+ "Task": "Syntax (STX)",
4143
+ "Result": 79.57,
4144
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4145
+ "Params": 14,
4146
+ "Release": "V3",
4147
+ "Thinking": "Reasoning"
4148
+ },
4149
+ {
4150
+ "Model": "Hermes-4-14B-Reasoning",
4151
+ "Model Type": "General",
4152
+ "Benchmark": "VerilogEval S2R",
4153
+ "Task": "Functionality (FNC)",
4154
+ "Result": 56.15,
4155
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4156
+ "Params": 14,
4157
+ "Release": "V3",
4158
+ "Thinking": "Reasoning"
4159
+ },
4160
+ {
4161
+ "Model": "Hermes-4-14B-Reasoning",
4162
+ "Model Type": "General",
4163
+ "Benchmark": "RTLLM",
4164
+ "Task": "Functionality (FNC)",
4165
+ "Result": 41.28,
4166
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4167
+ "Params": 14,
4168
+ "Release": "V3",
4169
+ "Thinking": "Reasoning"
4170
+ },
4171
+ {
4172
+ "Model": "Hermes-4-14B-Reasoning",
4173
+ "Model Type": "General",
4174
+ "Benchmark": "VerilogEval S2R",
4175
+ "Task": "Synthesis (SYN)",
4176
+ "Result": 55.77,
4177
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4178
+ "Params": 14,
4179
+ "Release": "V3",
4180
+ "Thinking": "Reasoning"
4181
+ },
4182
+ {
4183
+ "Model": "Hermes-4-14B-Reasoning",
4184
+ "Model Type": "General",
4185
+ "Benchmark": "RTLLM",
4186
+ "Task": "Synthesis (SYN)",
4187
+ "Result": 34.47,
4188
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4189
+ "Params": 14,
4190
+ "Release": "V3",
4191
+ "Thinking": "Reasoning"
4192
+ },
4193
+ {
4194
+ "Model": "Hermes-4-14B-Reasoning",
4195
+ "Model Type": "General",
4196
+ "Benchmark": "VerilogEval S2R",
4197
+ "Task": "Power",
4198
+ "Result": 55.85,
4199
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4200
+ "Params": 14,
4201
+ "Release": "V3",
4202
+ "Thinking": "Reasoning"
4203
+ },
4204
+ {
4205
+ "Model": "Hermes-4-14B-Reasoning",
4206
+ "Model Type": "General",
4207
+ "Benchmark": "RTLLM",
4208
+ "Task": "Power",
4209
+ "Result": 38.35,
4210
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4211
+ "Params": 14,
4212
+ "Release": "V3",
4213
+ "Thinking": "Reasoning"
4214
+ },
4215
+ {
4216
+ "Model": "Hermes-4-14B-Reasoning",
4217
+ "Model Type": "General",
4218
+ "Benchmark": "VerilogEval S2R",
4219
+ "Task": "Performance",
4220
+ "Result": 53.71,
4221
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4222
+ "Params": 14,
4223
+ "Release": "V3",
4224
+ "Thinking": "Reasoning"
4225
+ },
4226
+ {
4227
+ "Model": "Hermes-4-14B-Reasoning",
4228
+ "Model Type": "General",
4229
+ "Benchmark": "RTLLM",
4230
+ "Task": "Performance",
4231
+ "Result": 32.73,
4232
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4233
+ "Params": 14,
4234
+ "Release": "V3",
4235
+ "Thinking": "Reasoning"
4236
+ },
4237
+ {
4238
+ "Model": "Hermes-4-14B-Reasoning",
4239
+ "Model Type": "General",
4240
+ "Benchmark": "VerilogEval S2R",
4241
+ "Task": "Area",
4242
+ "Result": 55.65,
4243
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4244
+ "Params": 14,
4245
+ "Release": "V3",
4246
+ "Thinking": "Reasoning"
4247
+ },
4248
+ {
4249
+ "Model": "Hermes-4-14B-Reasoning",
4250
+ "Model Type": "General",
4251
+ "Benchmark": "RTLLM",
4252
+ "Task": "Area",
4253
+ "Result": 36.13,
4254
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4255
+ "Params": 14,
4256
+ "Release": "V3",
4257
+ "Thinking": "Reasoning"
4258
+ },
4259
+ {
4260
+ "Model": "Hermes-4-14B-Reasoning",
4261
+ "Model Type": "General",
4262
+ "Benchmark": "RTL-Repo",
4263
+ "Task": "Exact Matching (EM)",
4264
+ "Result": -1.0,
4265
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4266
+ "Params": 14,
4267
+ "Release": "V3",
4268
+ "Thinking": "Reasoning"
4269
+ },
4270
+ {
4271
+ "Model": "Hermes-4-14B-Reasoning",
4272
+ "Model Type": "General",
4273
+ "Benchmark": "VerilogEval MC",
4274
+ "Task": "Syntax (STX)",
4275
+ "Result": 52.18,
4276
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4277
+ "Params": 14,
4278
+ "Release": "V3",
4279
+ "Thinking": "Reasoning"
4280
+ },
4281
+ {
4282
+ "Model": "Hermes-4-14B-Reasoning",
4283
+ "Model Type": "General",
4284
+ "Benchmark": "VeriGen",
4285
+ "Task": "Syntax (STX)",
4286
+ "Result": 84.71,
4287
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4288
+ "Params": 14,
4289
+ "Release": "V3",
4290
+ "Thinking": "Reasoning"
4291
+ },
4292
+ {
4293
+ "Model": "Hermes-4-14B-Reasoning",
4294
+ "Model Type": "General",
4295
+ "Benchmark": "VerilogEval MC",
4296
+ "Task": "Functionality (FNC)",
4297
+ "Result": 36.92,
4298
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4299
+ "Params": 14,
4300
+ "Release": "V3",
4301
+ "Thinking": "Reasoning"
4302
+ },
4303
+ {
4304
+ "Model": "Hermes-4-14B-Reasoning",
4305
+ "Model Type": "General",
4306
+ "Benchmark": "VeriGen",
4307
+ "Task": "Functionality (FNC)",
4308
+ "Result": 60.0,
4309
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4310
+ "Params": 14,
4311
+ "Release": "V3",
4312
+ "Thinking": "Reasoning"
4313
+ },
4314
+ {
4315
+ "Model": "Hermes-4-14B-Reasoning",
4316
+ "Model Type": "General",
4317
+ "Benchmark": "VerilogEval MC",
4318
+ "Task": "Synthesis (SYN)",
4319
+ "Result": 36.67,
4320
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4321
+ "Params": 14,
4322
+ "Release": "V3",
4323
+ "Thinking": "Reasoning"
4324
+ },
4325
+ {
4326
+ "Model": "Hermes-4-14B-Reasoning",
4327
+ "Model Type": "General",
4328
+ "Benchmark": "VeriGen",
4329
+ "Task": "Synthesis (SYN)",
4330
+ "Result": 58.82,
4331
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4332
+ "Params": 14,
4333
+ "Release": "V3",
4334
+ "Thinking": "Reasoning"
4335
+ },
4336
+ {
4337
+ "Model": "Hermes-4-14B-Reasoning",
4338
+ "Model Type": "General",
4339
+ "Benchmark": "VerilogEval MC",
4340
+ "Task": "Power",
4341
+ "Result": 36.21,
4342
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4343
+ "Params": 14,
4344
+ "Release": "V3",
4345
+ "Thinking": "Reasoning"
4346
+ },
4347
+ {
4348
+ "Model": "Hermes-4-14B-Reasoning",
4349
+ "Model Type": "General",
4350
+ "Benchmark": "VeriGen",
4351
+ "Task": "Power",
4352
+ "Result": 52.97,
4353
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4354
+ "Params": 14,
4355
+ "Release": "V3",
4356
+ "Thinking": "Reasoning"
4357
+ },
4358
+ {
4359
+ "Model": "Hermes-4-14B-Reasoning",
4360
+ "Model Type": "General",
4361
+ "Benchmark": "VerilogEval MC",
4362
+ "Task": "Performance",
4363
+ "Result": 36.02,
4364
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4365
+ "Params": 14,
4366
+ "Release": "V3",
4367
+ "Thinking": "Reasoning"
4368
+ },
4369
+ {
4370
+ "Model": "Hermes-4-14B-Reasoning",
4371
+ "Model Type": "General",
4372
+ "Benchmark": "VeriGen",
4373
+ "Task": "Performance",
4374
+ "Result": 58.57,
4375
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4376
+ "Params": 14,
4377
+ "Release": "V3",
4378
+ "Thinking": "Reasoning"
4379
+ },
4380
+ {
4381
+ "Model": "Hermes-4-14B-Reasoning",
4382
+ "Model Type": "General",
4383
+ "Benchmark": "VerilogEval MC",
4384
+ "Task": "Area",
4385
+ "Result": 36.12,
4386
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4387
+ "Params": 14,
4388
+ "Release": "V3",
4389
+ "Thinking": "Reasoning"
4390
+ },
4391
+ {
4392
+ "Model": "Hermes-4-14B-Reasoning",
4393
+ "Model Type": "General",
4394
+ "Benchmark": "VeriGen",
4395
+ "Task": "Area",
4396
+ "Result": 59.8,
4397
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4398
+ "Params": 14,
4399
+ "Release": "V3",
4400
+ "Thinking": "Reasoning"
4401
+ },
4402
+ {
4403
+ "Model": "Hermes-4-14B",
4404
+ "Model Type": "General",
4405
+ "Benchmark": "VerilogEval S2R",
4406
+ "Task": "Syntax (STX)",
4407
+ "Result": 72.18,
4408
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4409
+ "Params": 14,
4410
+ "Release": "V3",
4411
+ "Thinking": "Dense"
4412
+ },
4413
+ {
4414
+ "Model": "Hermes-4-14B",
4415
+ "Model Type": "General",
4416
+ "Benchmark": "RTLLM",
4417
+ "Task": "Syntax (STX)",
4418
+ "Result": 85.11,
4419
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4420
+ "Params": 14,
4421
+ "Release": "V3",
4422
+ "Thinking": "Dense"
4423
+ },
4424
+ {
4425
+ "Model": "Hermes-4-14B",
4426
+ "Model Type": "General",
4427
+ "Benchmark": "VerilogEval S2R",
4428
+ "Task": "Functionality (FNC)",
4429
+ "Result": 43.85,
4430
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4431
+ "Params": 14,
4432
+ "Release": "V3",
4433
+ "Thinking": "Dense"
4434
+ },
4435
+ {
4436
+ "Model": "Hermes-4-14B",
4437
+ "Model Type": "General",
4438
+ "Benchmark": "RTLLM",
4439
+ "Task": "Functionality (FNC)",
4440
+ "Result": 50.64,
4441
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4442
+ "Params": 14,
4443
+ "Release": "V3",
4444
+ "Thinking": "Dense"
4445
+ },
4446
+ {
4447
+ "Model": "Hermes-4-14B",
4448
+ "Model Type": "General",
4449
+ "Benchmark": "VerilogEval S2R",
4450
+ "Task": "Synthesis (SYN)",
4451
+ "Result": 43.85,
4452
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4453
+ "Params": 14,
4454
+ "Release": "V3",
4455
+ "Thinking": "Dense"
4456
+ },
4457
+ {
4458
+ "Model": "Hermes-4-14B",
4459
+ "Model Type": "General",
4460
+ "Benchmark": "RTLLM",
4461
+ "Task": "Synthesis (SYN)",
4462
+ "Result": 45.96,
4463
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4464
+ "Params": 14,
4465
+ "Release": "V3",
4466
+ "Thinking": "Dense"
4467
+ },
4468
+ {
4469
+ "Model": "Hermes-4-14B",
4470
+ "Model Type": "General",
4471
+ "Benchmark": "VerilogEval S2R",
4472
+ "Task": "Power",
4473
+ "Result": 43.27,
4474
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4475
+ "Params": 14,
4476
+ "Release": "V3",
4477
+ "Thinking": "Dense"
4478
+ },
4479
+ {
4480
+ "Model": "Hermes-4-14B",
4481
+ "Model Type": "General",
4482
+ "Benchmark": "RTLLM",
4483
+ "Task": "Power",
4484
+ "Result": 51.01,
4485
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4486
+ "Params": 14,
4487
+ "Release": "V3",
4488
+ "Thinking": "Dense"
4489
+ },
4490
+ {
4491
+ "Model": "Hermes-4-14B",
4492
+ "Model Type": "General",
4493
+ "Benchmark": "VerilogEval S2R",
4494
+ "Task": "Performance",
4495
+ "Result": 41.87,
4496
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4497
+ "Params": 14,
4498
+ "Release": "V3",
4499
+ "Thinking": "Dense"
4500
+ },
4501
+ {
4502
+ "Model": "Hermes-4-14B",
4503
+ "Model Type": "General",
4504
+ "Benchmark": "RTLLM",
4505
+ "Task": "Performance",
4506
+ "Result": 46.23,
4507
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4508
+ "Params": 14,
4509
+ "Release": "V3",
4510
+ "Thinking": "Dense"
4511
+ },
4512
+ {
4513
+ "Model": "Hermes-4-14B",
4514
+ "Model Type": "General",
4515
+ "Benchmark": "VerilogEval S2R",
4516
+ "Task": "Area",
4517
+ "Result": 42.92,
4518
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4519
+ "Params": 14,
4520
+ "Release": "V3",
4521
+ "Thinking": "Dense"
4522
+ },
4523
+ {
4524
+ "Model": "Hermes-4-14B",
4525
+ "Model Type": "General",
4526
+ "Benchmark": "RTLLM",
4527
+ "Task": "Area",
4528
+ "Result": 48.33,
4529
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4530
+ "Params": 14,
4531
+ "Release": "V3",
4532
+ "Thinking": "Dense"
4533
+ },
4534
+ {
4535
+ "Model": "Hermes-4-14B",
4536
+ "Model Type": "General",
4537
+ "Benchmark": "RTL-Repo",
4538
+ "Task": "Exact Matching (EM)",
4539
+ "Result": 28.36,
4540
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4541
+ "Params": 14,
4542
+ "Release": "V3",
4543
+ "Thinking": "Dense"
4544
+ },
4545
+ {
4546
+ "Model": "Hermes-4-14B",
4547
+ "Model Type": "General",
4548
+ "Benchmark": "VerilogEval MC",
4549
+ "Task": "Syntax (STX)",
4550
+ "Result": 54.23,
4551
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4552
+ "Params": 14,
4553
+ "Release": "V3",
4554
+ "Thinking": "Dense"
4555
+ },
4556
+ {
4557
+ "Model": "Hermes-4-14B",
4558
+ "Model Type": "General",
4559
+ "Benchmark": "VeriGen",
4560
+ "Task": "Syntax (STX)",
4561
+ "Result": 84.71,
4562
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4563
+ "Params": 14,
4564
+ "Release": "V3",
4565
+ "Thinking": "Dense"
4566
+ },
4567
+ {
4568
+ "Model": "Hermes-4-14B",
4569
+ "Model Type": "General",
4570
+ "Benchmark": "VerilogEval MC",
4571
+ "Task": "Functionality (FNC)",
4572
+ "Result": 27.69,
4573
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4574
+ "Params": 14,
4575
+ "Release": "V3",
4576
+ "Thinking": "Dense"
4577
+ },
4578
+ {
4579
+ "Model": "Hermes-4-14B",
4580
+ "Model Type": "General",
4581
+ "Benchmark": "VeriGen",
4582
+ "Task": "Functionality (FNC)",
4583
+ "Result": 48.24,
4584
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4585
+ "Params": 14,
4586
+ "Release": "V3",
4587
+ "Thinking": "Dense"
4588
+ },
4589
+ {
4590
+ "Model": "Hermes-4-14B",
4591
+ "Model Type": "General",
4592
+ "Benchmark": "VerilogEval MC",
4593
+ "Task": "Synthesis (SYN)",
4594
+ "Result": 27.69,
4595
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4596
+ "Params": 14,
4597
+ "Release": "V3",
4598
+ "Thinking": "Dense"
4599
+ },
4600
+ {
4601
+ "Model": "Hermes-4-14B",
4602
+ "Model Type": "General",
4603
+ "Benchmark": "VeriGen",
4604
+ "Task": "Synthesis (SYN)",
4605
+ "Result": 48.24,
4606
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4607
+ "Params": 14,
4608
+ "Release": "V3",
4609
+ "Thinking": "Dense"
4610
+ },
4611
+ {
4612
+ "Model": "Hermes-4-14B",
4613
+ "Model Type": "General",
4614
+ "Benchmark": "VerilogEval MC",
4615
+ "Task": "Power",
4616
+ "Result": 27.36,
4617
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4618
+ "Params": 14,
4619
+ "Release": "V3",
4620
+ "Thinking": "Dense"
4621
+ },
4622
+ {
4623
+ "Model": "Hermes-4-14B",
4624
+ "Model Type": "General",
4625
+ "Benchmark": "VeriGen",
4626
+ "Task": "Power",
4627
+ "Result": 46.18,
4628
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4629
+ "Params": 14,
4630
+ "Release": "V3",
4631
+ "Thinking": "Dense"
4632
+ },
4633
+ {
4634
+ "Model": "Hermes-4-14B",
4635
+ "Model Type": "General",
4636
+ "Benchmark": "VerilogEval MC",
4637
+ "Task": "Performance",
4638
+ "Result": 27.34,
4639
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4640
+ "Params": 14,
4641
+ "Release": "V3",
4642
+ "Thinking": "Dense"
4643
+ },
4644
+ {
4645
+ "Model": "Hermes-4-14B",
4646
+ "Model Type": "General",
4647
+ "Benchmark": "VeriGen",
4648
+ "Task": "Performance",
4649
+ "Result": 48.57,
4650
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4651
+ "Params": 14,
4652
+ "Release": "V3",
4653
+ "Thinking": "Dense"
4654
+ },
4655
+ {
4656
+ "Model": "Hermes-4-14B",
4657
+ "Model Type": "General",
4658
+ "Benchmark": "VerilogEval MC",
4659
+ "Task": "Area",
4660
+ "Result": 26.2,
4661
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4662
+ "Params": 14,
4663
+ "Release": "V3",
4664
+ "Thinking": "Dense"
4665
+ },
4666
+ {
4667
+ "Model": "Hermes-4-14B",
4668
+ "Model Type": "General",
4669
+ "Benchmark": "VeriGen",
4670
+ "Task": "Area",
4671
+ "Result": 47.47,
4672
+ "Model URL": "https://huggingface.co/NousResearch/Hermes-4-14B",
4673
+ "Params": 14,
4674
+ "Release": "V3",
4675
+ "Thinking": "Dense"
4676
+ },
4677
+ {
4678
+ "Model": "Qwen3-8B",
4679
+ "Model Type": "General",
4680
+ "Benchmark": "VerilogEval S2R",
4681
+ "Task": "Syntax (STX)",
4682
+ "Result": 70.9,
4683
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4684
+ "Params": 8.2,
4685
+ "Release": "V3",
4686
+ "Thinking": "Reasoning"
4687
+ },
4688
+ {
4689
+ "Model": "Qwen3-8B",
4690
+ "Model Type": "General",
4691
+ "Benchmark": "RTLLM",
4692
+ "Task": "Syntax (STX)",
4693
+ "Result": 65.96,
4694
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4695
+ "Params": 8.2,
4696
+ "Release": "V3",
4697
+ "Thinking": "Reasoning"
4698
+ },
4699
+ {
4700
+ "Model": "Qwen3-8B",
4701
+ "Model Type": "General",
4702
+ "Benchmark": "VerilogEval S2R",
4703
+ "Task": "Functionality (FNC)",
4704
+ "Result": 51.03,
4705
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4706
+ "Params": 8.2,
4707
+ "Release": "V3",
4708
+ "Thinking": "Reasoning"
4709
+ },
4710
+ {
4711
+ "Model": "Qwen3-8B",
4712
+ "Model Type": "General",
4713
+ "Benchmark": "RTLLM",
4714
+ "Task": "Functionality (FNC)",
4715
+ "Result": 37.87,
4716
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4717
+ "Params": 8.2,
4718
+ "Release": "V3",
4719
+ "Thinking": "Reasoning"
4720
+ },
4721
+ {
4722
+ "Model": "Qwen3-8B",
4723
+ "Model Type": "General",
4724
+ "Benchmark": "VerilogEval S2R",
4725
+ "Task": "Synthesis (SYN)",
4726
+ "Result": 51.03,
4727
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4728
+ "Params": 8.2,
4729
+ "Release": "V3",
4730
+ "Thinking": "Reasoning"
4731
+ },
4732
+ {
4733
+ "Model": "Qwen3-8B",
4734
+ "Model Type": "General",
4735
+ "Benchmark": "RTLLM",
4736
+ "Task": "Synthesis (SYN)",
4737
+ "Result": 33.62,
4738
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4739
+ "Params": 8.2,
4740
+ "Release": "V3",
4741
+ "Thinking": "Reasoning"
4742
+ },
4743
+ {
4744
+ "Model": "Qwen3-8B",
4745
+ "Model Type": "General",
4746
+ "Benchmark": "VerilogEval S2R",
4747
+ "Task": "Power",
4748
+ "Result": 50.57,
4749
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4750
+ "Params": 8.2,
4751
+ "Release": "V3",
4752
+ "Thinking": "Reasoning"
4753
+ },
4754
+ {
4755
+ "Model": "Qwen3-8B",
4756
+ "Model Type": "General",
4757
+ "Benchmark": "RTLLM",
4758
+ "Task": "Power",
4759
+ "Result": 36.52,
4760
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4761
+ "Params": 8.2,
4762
+ "Release": "V3",
4763
+ "Thinking": "Reasoning"
4764
+ },
4765
+ {
4766
+ "Model": "Qwen3-8B",
4767
+ "Model Type": "General",
4768
+ "Benchmark": "VerilogEval S2R",
4769
+ "Task": "Performance",
4770
+ "Result": 48.9,
4771
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4772
+ "Params": 8.2,
4773
+ "Release": "V3",
4774
+ "Thinking": "Reasoning"
4775
+ },
4776
+ {
4777
+ "Model": "Qwen3-8B",
4778
+ "Model Type": "General",
4779
+ "Benchmark": "RTLLM",
4780
+ "Task": "Performance",
4781
+ "Result": 33.6,
4782
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4783
+ "Params": 8.2,
4784
+ "Release": "V3",
4785
+ "Thinking": "Reasoning"
4786
+ },
4787
+ {
4788
+ "Model": "Qwen3-8B",
4789
+ "Model Type": "General",
4790
+ "Benchmark": "VerilogEval S2R",
4791
+ "Task": "Area",
4792
+ "Result": 49.93,
4793
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4794
+ "Params": 8.2,
4795
+ "Release": "V3",
4796
+ "Thinking": "Reasoning"
4797
+ },
4798
+ {
4799
+ "Model": "Qwen3-8B",
4800
+ "Model Type": "General",
4801
+ "Benchmark": "RTLLM",
4802
+ "Task": "Area",
4803
+ "Result": 33.2,
4804
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4805
+ "Params": 8.2,
4806
+ "Release": "V3",
4807
+ "Thinking": "Reasoning"
4808
+ },
4809
+ {
4810
+ "Model": "Qwen3-8B",
4811
+ "Model Type": "General",
4812
+ "Benchmark": "RTL-Repo",
4813
+ "Task": "Exact Matching (EM)",
4814
+ "Result": -1.0,
4815
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4816
+ "Params": 8.2,
4817
+ "Release": "V3",
4818
+ "Thinking": "Reasoning"
4819
+ },
4820
+ {
4821
+ "Model": "Qwen3-8B",
4822
+ "Model Type": "General",
4823
+ "Benchmark": "VerilogEval MC",
4824
+ "Task": "Syntax (STX)",
4825
+ "Result": 70.77,
4826
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4827
+ "Params": 8.2,
4828
+ "Release": "V3",
4829
+ "Thinking": "Reasoning"
4830
+ },
4831
+ {
4832
+ "Model": "Qwen3-8B",
4833
+ "Model Type": "General",
4834
+ "Benchmark": "VeriGen",
4835
+ "Task": "Syntax (STX)",
4836
+ "Result": 83.53,
4837
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4838
+ "Params": 8.2,
4839
+ "Release": "V3",
4840
+ "Thinking": "Reasoning"
4841
+ },
4842
+ {
4843
+ "Model": "Qwen3-8B",
4844
+ "Model Type": "General",
4845
+ "Benchmark": "VerilogEval MC",
4846
+ "Task": "Functionality (FNC)",
4847
+ "Result": 51.41,
4848
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4849
+ "Params": 8.2,
4850
+ "Release": "V3",
4851
+ "Thinking": "Reasoning"
4852
+ },
4853
+ {
4854
+ "Model": "Qwen3-8B",
4855
+ "Model Type": "General",
4856
+ "Benchmark": "VeriGen",
4857
+ "Task": "Functionality (FNC)",
4858
+ "Result": 40.0,
4859
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4860
+ "Params": 8.2,
4861
+ "Release": "V3",
4862
+ "Thinking": "Reasoning"
4863
+ },
4864
+ {
4865
+ "Model": "Qwen3-8B",
4866
+ "Model Type": "General",
4867
+ "Benchmark": "VerilogEval MC",
4868
+ "Task": "Synthesis (SYN)",
4869
+ "Result": 51.15,
4870
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4871
+ "Params": 8.2,
4872
+ "Release": "V3",
4873
+ "Thinking": "Reasoning"
4874
+ },
4875
+ {
4876
+ "Model": "Qwen3-8B",
4877
+ "Model Type": "General",
4878
+ "Benchmark": "VeriGen",
4879
+ "Task": "Synthesis (SYN)",
4880
+ "Result": 38.82,
4881
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4882
+ "Params": 8.2,
4883
+ "Release": "V3",
4884
+ "Thinking": "Reasoning"
4885
+ },
4886
+ {
4887
+ "Model": "Qwen3-8B",
4888
+ "Model Type": "General",
4889
+ "Benchmark": "VerilogEval MC",
4890
+ "Task": "Power",
4891
+ "Result": 50.58,
4892
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4893
+ "Params": 8.2,
4894
+ "Release": "V3",
4895
+ "Thinking": "Reasoning"
4896
+ },
4897
+ {
4898
+ "Model": "Qwen3-8B",
4899
+ "Model Type": "General",
4900
+ "Benchmark": "VeriGen",
4901
+ "Task": "Power",
4902
+ "Result": 37.47,
4903
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4904
+ "Params": 8.2,
4905
+ "Release": "V3",
4906
+ "Thinking": "Reasoning"
4907
+ },
4908
+ {
4909
+ "Model": "Qwen3-8B",
4910
+ "Model Type": "General",
4911
+ "Benchmark": "VerilogEval MC",
4912
+ "Task": "Performance",
4913
+ "Result": 49.17,
4914
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4915
+ "Params": 8.2,
4916
+ "Release": "V3",
4917
+ "Thinking": "Reasoning"
4918
+ },
4919
+ {
4920
+ "Model": "Qwen3-8B",
4921
+ "Model Type": "General",
4922
+ "Benchmark": "VeriGen",
4923
+ "Task": "Performance",
4924
+ "Result": 38.76,
4925
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4926
+ "Params": 8.2,
4927
+ "Release": "V3",
4928
+ "Thinking": "Reasoning"
4929
+ },
4930
+ {
4931
+ "Model": "Qwen3-8B",
4932
+ "Model Type": "General",
4933
+ "Benchmark": "VerilogEval MC",
4934
+ "Task": "Area",
4935
+ "Result": 50.18,
4936
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4937
+ "Params": 8.2,
4938
+ "Release": "V3",
4939
+ "Thinking": "Reasoning"
4940
+ },
4941
+ {
4942
+ "Model": "Qwen3-8B",
4943
+ "Model Type": "General",
4944
+ "Benchmark": "VeriGen",
4945
+ "Task": "Area",
4946
+ "Result": 38.73,
4947
+ "Model URL": "https://huggingface.co/Qwen/Qwen3-8B",
4948
+ "Params": 8.2,
4949
+ "Release": "V3",
4950
+ "Thinking": "Reasoning"
4951
+ },
4952
  {
4953
  "Model": "CodeLlama 70B",
4954
  "Model Type": "Coding",
results/results_verilator_october_2025.csv CHANGED
@@ -33,8 +33,8 @@ RTLCoder DeepSeek,84.739999999999995,76.170000000000002,40.509999999999998,42.13
33
  OriGen,96.790000000000006,85.109999999999999,54.359999999999999,53.189999999999998,53.850000000000001,51.060000000000002,54.079999999999998,54.170000000000002,51.07,51.310000000000002,53.009999999999998,54.390000000000001,17.07,92.560000000000002,96.469999999999999,50.509999999999998,74.120000000000005,50.509999999999998,74.120000000000005,50.829999999999998,69.299999999999997,48.280000000000001,74.260000000000005,50.409999999999997,76.349999999999994
34
  CodeV-CL-7B,32.82,54.469999999999999,13.460000000000001,29.359999999999999,13.33,22.550000000000001,13.19,23.190000000000001,13.26,21.469999999999999,13.19,22.190000000000001,12.27,92.180000000000007,98.819999999999993,31.789999999999999,49.409999999999997,31.789999999999999,49.409999999999997,31.739999999999998,48.130000000000003,29.829999999999998,49.340000000000003,31.609999999999999,49.079999999999998
35
  CodeV-DS-6.7B,34.359999999999999,69.790000000000006,14.74,41.280000000000001,14.359999999999999,37.869999999999997,14.380000000000001,35.890000000000001,14.130000000000001,35.740000000000002,14.19,36.18,21.260000000000002,95.769999999999996,100,46.670000000000002,61.18,46.670000000000002,61.18,47.07,61.060000000000002,44.189999999999998,61.130000000000003,46.469999999999999,60.950000000000003
36
- ,,,,,,,,,,,,,,,,,,,,,,,,,
37
- ,,,,,,,,,,,,,,,,,,,,,,,,,
38
  ,,,,,,,,,,,,,,,,,,,,,,,,,
39
  ,,,,,,,,,,,,,,,,,,,,,,,,,
40
  ,,,,,,,,,,,,,,,,,,,,,,,,,
 
33
  OriGen,96.790000000000006,85.109999999999999,54.359999999999999,53.189999999999998,53.850000000000001,51.060000000000002,54.079999999999998,54.170000000000002,51.07,51.310000000000002,53.009999999999998,54.390000000000001,17.07,92.560000000000002,96.469999999999999,50.509999999999998,74.120000000000005,50.509999999999998,74.120000000000005,50.829999999999998,69.299999999999997,48.280000000000001,74.260000000000005,50.409999999999997,76.349999999999994
34
  CodeV-CL-7B,32.82,54.469999999999999,13.460000000000001,29.359999999999999,13.33,22.550000000000001,13.19,23.190000000000001,13.26,21.469999999999999,13.19,22.190000000000001,12.27,92.180000000000007,98.819999999999993,31.789999999999999,49.409999999999997,31.789999999999999,49.409999999999997,31.739999999999998,48.130000000000003,29.829999999999998,49.340000000000003,31.609999999999999,49.079999999999998
35
  CodeV-DS-6.7B,34.359999999999999,69.790000000000006,14.74,41.280000000000001,14.359999999999999,37.869999999999997,14.380000000000001,35.890000000000001,14.130000000000001,35.740000000000002,14.19,36.18,21.260000000000002,95.769999999999996,100,46.670000000000002,61.18,46.670000000000002,61.18,47.07,61.060000000000002,44.189999999999998,61.130000000000003,46.469999999999999,60.950000000000003
36
+ Qwen3-8B,70.900000000000006,65.959999999999994,51.030000000000001,37.869999999999997,51.030000000000001,33.619999999999997,50.57,36.520000000000003,48.899999999999999,33.600000000000001,49.93,33.200000000000003,-1,70.769999999999996,83.530000000000001,51.409999999999997,40,51.149999999999999,38.82,50.579999999999998,37.469999999999999,49.170000000000002,38.759999999999998,50.18,38.729999999999997
37
+ Hermes-4-14B,72.180000000000007,85.109999999999999,43.850000000000001,50.640000000000001,43.850000000000001,45.960000000000001,43.270000000000003,51.009999999999998,41.869999999999997,46.229999999999997,42.920000000000002,48.329999999999998,-1,54.229999999999997,84.709999999999994,27.690000000000001,48.240000000000002,27.690000000000001,48.240000000000002,27.359999999999999,46.18,27.34,48.57,26.199999999999999,47.469999999999999
38
  ,,,,,,,,,,,,,,,,,,,,,,,,,
39
  ,,,,,,,,,,,,,,,,,,,,,,,,,
40
  ,,,,,,,,,,,,,,,,,,,,,,,,,
static/html_content.py CHANGED
@@ -44,6 +44,9 @@ INTRO_HTML = """
44
  Use the filters below to explore different RTL benchmarks, simulators and models.
45
  </p>
46
  <p style="margin-top:10px; text-align:start !important;">
 
 
 
47
  <span style="font-variant:small-caps; font-weight:bold;">UPDATE (SEPT 2025):</span> Added <span>gpt-oss-20b</span> and <span>gpt-oss-120b</span> to the leaderboard
48
  </p>
49
  <p style="margin-top:-6px; text-align:start !important;">
@@ -105,3 +108,11 @@ REFERENCES_HTML = """
105
  </ul>
106
  </div>
107
  """
 
 
 
 
 
 
 
 
 
44
  Use the filters below to explore different RTL benchmarks, simulators and models.
45
  </p>
46
  <p style="margin-top:10px; text-align:start !important;">
47
+ <span style="font-variant:small-caps; font-weight:bold;">UPDATE (OCT 2025):</span> Added <span>Hermes-4-14B</span>, <span>Qwen3-8B</span>, and <span>Seed-OSS-36B</span> to the leaderboard. Implemented Other Models tab and moved models to it
48
+ </p>
49
+ <p style="margin-top:-6px; text-align:start !important;">
50
  <span style="font-variant:small-caps; font-weight:bold;">UPDATE (SEPT 2025):</span> Added <span>gpt-oss-20b</span> and <span>gpt-oss-120b</span> to the leaderboard
51
  </p>
52
  <p style="margin-top:-6px; text-align:start !important;">
 
108
  </ul>
109
  </div>
110
  """
111
+
112
+ OTHER_MODELS_HTML = """
113
+ <div style="max-width: 800px; margin: auto; padding: 20px; border: 1px solid #ccc; border-radius: 10px;">
114
+ <p style="font-size: 16px; text-align: start;">
115
+ This models were previously listed on the main leaderboard and will no longer be updated.
116
+ </p>
117
+ </div>
118
+ """