Benchmark
GPQA Diamond
Science benchmarks measure how well models answer science questions. Scores are percentages. Higher is better.
- Models and systems measured
- 282
- Scores recorded
- 1623
- Sources
- 4
- IQ weight
- 1.72
Score distribution
- 79.3%
- Mean
- 84.5%
- Median
- 15.7%
- Std dev
- 56.8%
- 10th percentile
- 92.2%
- 90th percentile
- 0.168%
- Min
This benchmark has a weight of 1.72 for chat models, measured over 272 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 | |
|---|---|---|
| 1401 | DeepSeek V3OpenRouter harness, deepseek/deepseek-chat-v3-0324, Crusoe, public 90-day aggregate· OpenRouter benchmark exports | 63.6% |
| 1402 | Qwen3 30B A3B InstructOpenRouter harness, qwen/qwen3-30b-a3b-instruct-2507, CoreWeave, public 90-day aggregate· OpenRouter benchmark exports | 63.6% |
| 1403 | GPT-5.6 Lunanone· Epoch AI Benchmarking Hub data | 63.6% |
| 1404 | Qwen3 30B A3B InstructOpenRouter harness, qwen/qwen3-30b-a3b-instruct-2507, SiliconFlow, public 90-day aggregate· OpenRouter benchmark exports | 63.6% |
| 1405 | Nemotron 3 Nano 30B A3BOpenRouter harness, nvidia/nemotron-3-nano-30b-a3b, Crusoe, public 90-day aggregate· OpenRouter benchmark exports | 63.5% |
| 1406 | Nemotron 3.5 LightningOpenRouter harness, nvidia/nemotron-3.5-lightning-20260807, Fireworks, public 90-day aggregate· OpenRouter benchmark exports | 63.4% |
| 1407 | Trinity Large PreviewTrinity technical report · Table 4 · exact reported precision· Reported by Arcee AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 63.3% |
| 1408 | Trinity Large PreviewTrinity-Large-Preview Model Card · rounded to one decimal· Reported by Arcee AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 63.3% |
| 1409 | Ling 3.0 Flash FinOpenRouter harness, inclusionai/ling-3.0-flash-fin-20260827, Novita, public 90-day aggregate· OpenRouter benchmark exports | 63.2% |
| 1410 | Nemotron 3 Nano 30B A3BOpenRouter harness, nvidia/nemotron-3-nano-30b-a3b, auto-routing, public 90-day aggregate· OpenRouter benchmark exports | 63.1% |
| 1411 | Qwen3 30B A3B InstructOpenRouter harness, qwen/qwen3-30b-a3b-instruct-2507, StreamLake, public 90-day aggregate· OpenRouter benchmark exports | 63.1% |
| 1412 | Qwen3 30B A3B InstructOpenRouter harness, qwen/qwen3-30b-a3b-instruct-2507, DekaLLM, public 90-day aggregate· OpenRouter benchmark exports | 63.1% |
| 1413 | GPT OSS 20bOpenRouter harness, openai/gpt-oss-20b, SiliconFlow, public 90-day aggregate· OpenRouter benchmark exports | 63% |
| 1414 | Nemotron 3 Nano 30B A3BOpenRouter harness, nvidia/nemotron-3-nano-30b-a3b, Nebius, public 90-day aggregate· OpenRouter benchmark exports | 62.7% |
| 1415 | Gemma 4 26B A4BOpenRouter harness, google/gemma-4-26b-a4b-it-20260403, Darkbloom, public 90-day aggregate· OpenRouter benchmark exports | 62.7% |
| 1416 | DeepSeek V3.1OpenRouter harness, deepseek/deepseek-chat-v3.1, auto-routing, public 90-day aggregate· OpenRouter benchmark exports | 62.5% |
| 1417 | MiMo V2.5 ProOpenRouter harness, xiaomi/mimo-v2.5-pro-20260422, AtlasCloud, public 90-day aggregate· OpenRouter benchmark exports | 62.5% |
| 1418 | O1 Minihigh· Epoch AI Benchmarking Hub data | 62.4% |
| 1419 | DeepSeek ChatOpenRouter harness, deepseek/deepseek-chat-v3, auto-routing, public 90-day aggregate· OpenRouter benchmark exports | 62.1% |
| 1420 | Qwen3 30B A3B InstructOpenRouter harness, qwen/qwen3-30b-a3b-instruct-2507, PrimeIntellect, public 90-day aggregate· OpenRouter benchmark exports | 62.1% |
| 1421 | Qwen3 32BOpenRouter harness, qwen/qwen3-32b-04-28, Nebius, public 90-day aggregate· OpenRouter benchmark exports | 62.1% |
| 1422 | Qwen3 30B A3B InstructOpenRouter harness, qwen/qwen3-30b-a3b-instruct-2507, Nebius, public 90-day aggregate· OpenRouter benchmark exports | 62.1% |
| 1423 | GLM 4.7 FlashOpenRouter harness, z-ai/glm-4.7-flash-20260119, Cloudflare, public 90-day aggregate· OpenRouter benchmark exports | 62% |
| 1424 | Qwen3 4B Instruct 2507· OpenEvals Leaderboard data· Reported by the model's developer. The source relaying it did not verify it.Values from 2 sources
| 62% |
| 1425 | GPT OSS 120BOpenRouter harness, openai/gpt-oss-120b, Phala, public 90-day aggregate· OpenRouter benchmark exports | 62% |
| 1426 | GPT OSS 20bOpenRouter harness, openai/gpt-oss-20b, Groq, public 90-day aggregate· OpenRouter benchmark exports | 62% |
| 1427 | Qwen3 30B A3B· Epoch AI Benchmarking Hub data | 61.7% |
| 1428 | Qwen3 CoderOpenRouter harness, qwen/qwen3-coder-480b-a35b-07-25, Novita, public 90-day aggregate· OpenRouter benchmark exports | 61.5% |
| 1429 | Qwen3 14BOpenRouter harness, qwen/qwen3-14b-04-28, DeepInfra, public 90-day aggregate· OpenRouter benchmark exports | 61.2% |
| 1430 | Qwen3 CoderOpenRouter harness, qwen/qwen3-coder-480b-a35b-07-25, Google, public 90-day aggregate· OpenRouter benchmark exports | 61% |
| 1431 | Qwen3 235B A22B Instruct 2507OpenRouter harness, qwen/qwen3-235b-a22b-07-25, AtlasCloud, public 90-day aggregate· OpenRouter benchmark exports | 60.9% |
| 1432 | Qwen3 30B A3BOpenRouter harness, qwen/qwen3-30b-a3b-04-28, DeepInfra, public 90-day aggregate· OpenRouter benchmark exports | 60.9% |
| 1433 | GLM 4.7OpenRouter harness, z-ai/glm-4.7-20251222, Mancer 2, public 90-day aggregate· OpenRouter benchmark exports | 60.8% |
| 1434 | GPT OSS 20bmedium· Epoch AI Benchmarking Hub data | 60.8% |
| 1435 | qwen3-vl-30bOpenRouter harness, qwen/qwen3-vl-30b-a3b-instruct, SiliconFlow, public 90-day aggregate· OpenRouter benchmark exports | 60.8% |
| 1436 | Qwen3 CoderOpenRouter harness, qwen/qwen3-coder-480b-a35b-07-25, auto-routing, public 90-day aggregate· OpenRouter benchmark exports | 60.7% |
| 1437 | DeepSeek V3.2OpenRouter harness, deepseek/deepseek-v3.2-20251201, Mara, public 90-day aggregate· OpenRouter benchmark exports | 60.6% |
| 1438 | Qwen3 CoderOpenRouter harness, qwen/qwen3-coder-480b-a35b-07-25, DeepInfra, public 90-day aggregate· OpenRouter benchmark exports | 60.6% |
| 1439 | Claude Haiku 4.5· Epoch AI Benchmarking Hub data | 60.5% |
| 1440 | qwen3-vl-30bOpenRouter harness, qwen/qwen3-vl-30b-a3b-instruct, DeepInfra, public 90-day aggregate· OpenRouter benchmark exports | 60.1% |
| 1441 | Qwen3 CoderOpenRouter harness, qwen/qwen3-coder-480b-a35b-07-25, Venice, public 90-day aggregate· OpenRouter benchmark exports | 60% |
| 1442 | Qwen3 CoderOpenRouter harness, qwen/qwen3-coder-480b-a35b-07-25, Alibaba, public 90-day aggregate· OpenRouter benchmark exports | 60% |
| 1443 | GPT OSS 20bOpenRouter harness, openai/gpt-oss-20b, PrimeIntellect, public 90-day aggregate· OpenRouter benchmark exports | 59.9% |
| 1444 | Qwen3 32BOpenRouter harness, qwen/qwen3-32b-04-28, auto-routing, public 90-day aggregate· OpenRouter benchmark exports | 59.9% |
| 1445 | Qwen3 14BOpenRouter harness, qwen/qwen3-14b-04-28, NextBit, public 90-day aggregate· OpenRouter benchmark exports | 59.8% |
| 1446 | Gemini 2.5 Flash-LiteOpenRouter harness, google/gemini-2.5-flash-lite, Google AI Studio, public 90-day aggregate· OpenRouter benchmark exports | 59.6% |
| 1447 | Nemotron 3 Nano 30B A3BOpenRouter harness, nvidia/nemotron-3-nano-30b-a3b, Novita, public 90-day aggregate· OpenRouter benchmark exports | 59.6% |
| 1448 | Mistral Medium 3· Epoch AI Benchmarking Hub data | 59.5% |
| 1449 | O1 Minimedium· Epoch AI Benchmarking Hub data | 59.5% |
| 1450 | DeepSeek V3.2OpenRouter harness, deepseek/deepseek-v3.2-20251201, SambaNova, public 90-day aggregate· OpenRouter benchmark exports | 59.3% |
Showing 1,401–1,450 of 1,621 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 | |
|---|---|
| MiMo V2 Flash Baseofficial Base table · 5-shot· Reported by Xiaomi MiMo· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 55.1% |
| Qwen3.8-2.4T-A95Blifecycle-page core-benchmark claim· Reported by Alibaba Cloud Model Studio· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it. | 92.6% |
Tokens the model generated per task, as measured by the source. Comparable within this benchmark only.
Data credits: Epoch AI Benchmarking Hub, CC BY 4.0 (licence) · 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. · OpenRouter public benchmark leaderboard and search-lane exports, CC BY 4.0. Fractional source scores converted to percentages; immutable dated snapshots. (licence) · OpenEvals/leaderboard-data, MIT (licence)