Skip to content

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
0.168%96.8%
The band covers the middle 80% of scores. The line marks the median.

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
1301Claude Haiku 4.532k· Epoch AI Benchmarking Hub data71.2%
1302DeepSeek Chat· Epoch AI Benchmarking Hub data71.2%
1303GLM 5.2none· Epoch AI Benchmarking Hub data71.2%
1304DeepSeek V4 FlashDeepSeek V4 official model card · Non-Think· Reported by DeepSeek· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.71.2%
1305Qwen3 VL 235BOpenRouter harness, qwen/qwen3-vl-235b-a22b-instruct, Alibaba, public 90-day aggregate· OpenRouter benchmark exports71.1%
1306Nemotron 3.5 LightningOpenRouter harness, nvidia/nemotron-3.5-lightning-20260807, DeepInfra, public 90-day aggregate· OpenRouter benchmark exports71.1%
1307Qwen3 Next 80B A3B InstructOpenRouter harness, qwen/qwen3-next-80b-a3b-instruct-2509, Novita, public 90-day aggregate· OpenRouter benchmark exports71%
1308Gemma 4 26B A4BOpenRouter harness, google/gemma-4-26b-a4b-it-20260403, AkashML, public 90-day aggregate· OpenRouter benchmark exports71%
1309Qwen3 VL 235BOpenRouter harness, qwen/qwen3-vl-235b-a22b-instruct, PrimeIntellect, public 90-day aggregate· OpenRouter benchmark exports70.7%
1310Qwen 3 235B A22B· Epoch AI Benchmarking Hub data70.7%
1311Qwen3 VL 235BOpenRouter harness, qwen/qwen3-vl-235b-a22b-instruct, Novita, public 90-day aggregate· OpenRouter benchmark exports70.6%
1312Nemotron 3.5 LightningOpenRouter harness, nvidia/nemotron-3.5-lightning-20260807, auto-routing, public 90-day aggregate· OpenRouter benchmark exports70.6%
1313Gemma 4 26B A4BOpenRouter harness, google/gemma-4-26b-a4b-it-20260403, auto-routing, public 90-day aggregate· OpenRouter benchmark exports70.6%
1314Qwen3 Next 80B A3B InstructOpenRouter harness, qwen/qwen3-next-80b-a3b-instruct-2509, DeepInfra, public 90-day aggregate· OpenRouter benchmark exports70.5%
1315GLM 4.6OpenRouter harness, z-ai/glm-4.6, Venice, public 90-day aggregate· OpenRouter benchmark exports70.5%
1316GPT-5 nanoOpenRouter harness, openai/gpt-5-nano-2025-08-07, auto-routing, public 90-day aggregate· OpenRouter benchmark exports70.5%
1317DeepSeek V4 ProOpenRouter harness, deepseek/deepseek-v4-pro-20260423, Venice, public 90-day aggregate· OpenRouter benchmark exports70.4%
1318GLM 4.7OpenRouter harness, z-ai/glm-4.7-20251222, Venice, public 90-day aggregate· OpenRouter benchmark exports70.3%
1319Qwen3.5 9BOpenRouter harness, qwen/qwen3.5-9b-20260310, Darkbloom, public 90-day aggregate· OpenRouter benchmark exports70.2%
1320Qwen3 30B A3B2507, thinking· Epoch AI Benchmarking Hub data70.1%
1321GPT OSS 120BOpenRouter harness, openai/gpt-oss-120b, Google, public 90-day aggregate· OpenRouter benchmark exports70%
1322Claude Sonnet 4no extended thinking· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.70%
1323Qwen3.6 35B A3BOpenRouter harness, qwen/qwen3.6-35b-a3b-20260415, Io Net, public 90-day aggregate· OpenRouter benchmark exports69.6%
1324Qwen3 VL 235BOpenRouter harness, qwen/qwen3-vl-235b-a22b-instruct, auto-routing, public 90-day aggregate· OpenRouter benchmark exports69.4%
1325GPT-5 nanohigh· Epoch AI Benchmarking Hub data69.4%
1326Claude Opus 4· Epoch AI Benchmarking Hub data69.2%
1327Qwen3 VL 235BOpenRouter harness, qwen/qwen3-vl-235b-a22b-instruct, DeepInfra, public 90-day aggregate· OpenRouter benchmark exports69.1%
1328Gemma 4 31B ITOpenRouter harness, google/gemma-4-31b-it-20260402, Friendli, public 90-day aggregate· OpenRouter benchmark exports69.1%
1329GPT OSS 120BOpenRouter harness, openai/gpt-oss-120b, SiliconFlow, public 90-day aggregate· OpenRouter benchmark exports69.1%
1330GPT-5 nanoOpenRouter harness, openai/gpt-5-nano-2025-08-07, Azure, public 90-day aggregate· OpenRouter benchmark exports69%
1331GPT-4.5 Preview· Epoch AI Benchmarking Hub data68.7%
1332GPT OSS 20bOpenRouter harness, openai/gpt-oss-20b, DekaLLM, public 90-day aggregate· OpenRouter benchmark exports68.7%
1333GPT-5 nanoOpenRouter harness, openai/gpt-5-nano-2025-08-07, OpenAI, public 90-day aggregate· OpenRouter benchmark exports68.4%
1334DeepSeek V3.1OpenRouter harness, deepseek/deepseek-chat-v3.1, SambaNova, public 90-day aggregate· OpenRouter benchmark exports68.3%
1335o3 Minilow· Epoch AI Benchmarking Hub data68.2%
1336GPT OSS 120BOpenRouter harness, openai/gpt-oss-120b, Mancer 2, public 90-day aggregate· OpenRouter benchmark exports68.2%
1337Gemma 4 31B ITOpenRouter harness, google/gemma-4-31b-it-20260402, OpenInference, public 90-day aggregate· OpenRouter benchmark exports68.2%
1338Nemotron 3 Nano 30B A3BOpenRouter harness, nvidia/nemotron-3-nano-30b-a3b, DeepInfra, public 90-day aggregate· OpenRouter benchmark exports68.1%
1339GPT OSS 120Bofficial model card Table 3 · reasoning low · with tools; Table 3 does not name the exact tool · pass@1· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.68.1%
1340MiniMax M2.5OpenRouter harness, minimax/minimax-m2.5-20260211, Venice, public 90-day aggregate· OpenRouter benchmark exports68.1%
1341GPT OSS 20bOpenRouter harness, openai/gpt-oss-20b, AkashML, public 90-day aggregate· OpenRouter benchmark exports67.8%
1342DeepSeek V3OpenRouter harness, deepseek/deepseek-chat-v3-0324, SiliconFlow, public 90-day aggregate· OpenRouter benchmark exports67.7%
1343DeepSeek V3· Epoch AI Benchmarking Hub data67.6%
1344Grok 3beta· Epoch AI Benchmarking Hub data67.6%
1345GPT OSS 20bOpenRouter harness, openai/gpt-oss-20b, DeepInfra, public 90-day aggregate· OpenRouter benchmark exports67.4%
1346GPT OSS 20bOpenRouter harness, openai/gpt-oss-20b, CoreWeave, public 90-day aggregate· OpenRouter benchmark exports67.4%
1347GPT-5 nanomedium· Epoch AI Benchmarking Hub data67.4%
1348GPT OSS 20bOpenRouter harness, openai/gpt-oss-20b, Parasail, public 90-day aggregate· OpenRouter benchmark exports67.1%
1349GPT OSS 120B· OpenEvals Leaderboard data· Reported by the model's developer. The source relaying it did not verify it.67.1%
1350GPT OSS 120Bofficial model card Table 3 · reasoning low · no tools · pass@1· Reported by OpenAI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.67.1%

Showing 1,301–1,350 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)