Benchmark
GPQA
Science benchmarks measure how well models answer science questions. Scores are percentages. Higher is better.
- Models and systems measured
- 364
- Scores recorded
- 378
- Sources
- 3
- IQ weight
- 0.75
Score distribution
- 45.3%
- Mean
- 47.5%
- Median
- 34.4%
- Std dev
- 2.9%
- 10th percentile
- 89%
- 90th percentile
- 0%
- Min
This benchmark has a weight of 0.75 for chat models, measured over 359 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model or system | Score |
|---|---|---|
| 201 | Mistral Small 24B Instruct2501· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 45.3% |
| 202 | Mistral Largelatest· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 43.9% |
| 203 | Gemma 4 E2B-it· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 43.4% |
| 204 | Gemma 3 27B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 42.4% |
| 205 | Qwen2 72B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 42.4% |
| 206 | Nova Lite· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 42% |
| 207 | Llama 3.1 70B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 41.7% |
| 208 | Claude Haiku 3.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 41.6% |
| 209 | Gemma 3 12B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 40.9% |
| 210 | Claude 3 Sonnet· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 40.4% |
| 211 | GPT-4o mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 40.2% |
| 212 | Nova Micro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 40% |
| 213 | Gemini 1.5 Flash-8B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 38.4% |
| 214 | Jamba 1.5 Large· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 36.9% |
| 215 | phi-3.5-moe-instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 36.8% |
| 216 | Qwen2.5 7B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 36.4% |
| 217 | GPT 4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 35.7% |
| 218 | DeepSeek R1 Distill Qwen 1.5B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 33.8% |
| 219 | Claude 3 Haiku· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 33.3% |
| 220 | Llama 3.2 11B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 32.8% |
| 221 | Llama 3.2 3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 32.8% |
| 222 | Jamba 1.5 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 32.3% |
| 223 | Gemma 3 4B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 30.8% |
| 224 | GPT-3.5 Turbo· LLM Stats data | 30.8% |
| 225 | Qwen2.5-Omni 7B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 30.8% |
| 226 | Llama 3.1 8B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 30.4% |
| 227 | L3.3 Ms Nevoria 70b· Hugging Face Open LLM Leaderboard data | 29.4% |
| 228 | L3.3 Nevoria R1 70b· Hugging Face Open LLM Leaderboard data | 29.2% |
| 229 | gemini-1.0-pro· LLM Stats data | 27.9% |
| 230 | Qwen2 7B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 25.3% |
| 231 | Phi 4 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claimValues from 2 sources
| 25.2% |
| 232 | Mistral Large Instruct 2411· Hugging Face Open LLM Leaderboard data | 24.9% |
| 233 | Gemma 3n E2b It· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 24.8% |
| 234 | Gemma 3n E4B IT· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 23.7% |
| 235 | Qwen2.5 32B· Hugging Face Open LLM Leaderboard data | 21.6% |
| 236 | EVA-Qwen2.5-72B-v0.2· Hugging Face Open LLM Leaderboard data | 21.1% |
| 237 | Qwen2.5 72B· Hugging Face Open LLM Leaderboard data | 20.7% |
| 238 | Meta Llama 3 70b· Hugging Face Open LLM Leaderboard data | 19.7% |
| 239 | Qwen 2 (72B)· Hugging Face Open LLM Leaderboard data | 19.2% |
| 240 | Gemma 3 1b it· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim | 19.2% |
| 241 | Qwen2 72B Dracarys· Hugging Face Open LLM Leaderboard data | 18.8% |
| 242 | Llama 3.1 70B· Hugging Face Open LLM Leaderboard data | 18.3% |
| 243 | Qwen2-VL (72B) Instruct· Hugging Face Open LLM Leaderboard data | 18.3% |
| 244 | Magnum V2 72B· Hugging Face Open LLM Leaderboard data | 18.1% |
| 245 | arcee-blitz· Hugging Face Open LLM Leaderboard data | 18% |
| 246 | Qwen2.5 14B· Hugging Face Open LLM Leaderboard data | 17.6% |
| 247 | WizardLM 2 8x22B· Hugging Face Open LLM Leaderboard data | 17.6% |
| 248 | mixtral-8x22b-v0.1· Hugging Face Open LLM Leaderboard data | 16.8% |
| 249 | Gemma 2 27B IT· Hugging Face Open LLM Leaderboard data | 16.7% |
| 250 | Mixtral 8x22B Instruct v0.1· Hugging Face Open LLM Leaderboard data | 16.4% |
Showing 201–250 of 374 results.
Not ranked
The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.
| Model or system | Score |
|---|---|
| Qwen3.5-397B-A17B-BaseQwen3.5 release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 54.6% |
| Qwen3.7-Plus-BaseQwen3.8-Flash-Next release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 51.5% |
| Qwen3.8-27B-BaseQwen3.8-Flash-Next release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 45% |
| Qwen3.8-Flash-Next-BaseQwen3.8-Flash-Next release/model card · base-model table· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 51.4% |
Data credits: Data by LLM Stats · BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.