Benchmark leaderboard - 20 entries
GPQA Diamond leaderboard
GPQA is a dataset of 448 multiple-choice questions written by domain experts in biology, physics, and chemistry, and GPQA Diamond is its 198-question subset where both expert validators answered correctly and at most one of three non-experts did. As of October 17, 2026, GPT-6 Astra is listed first on GPQA Diamond with a vendor-reported 96.0%. Every number here is reported by the model's own vendor under its own settings, so the order is not a controlled comparison.
Last verified from 8 sources. Read the caveats
GPQA Diamond
20 entries, in the board's order
| 1 | GPT-6 Astra OpenAI research environment or API; 'Evaluation scores are the maximum at any effort.' Vendor-reported | 96.0% |
| 2 | GPT-5.6 Sol OpenAI GPT-5.6 launch eval table (same value repeated in the GPT-6 Astra table) Vendor-reported | 94.6% |
| 3 | Claude Mythos Preview Anthropic system card: 'averaged over 5 trials' Vendor-reported | 94.55% |
| 4 | Gemini 3.1 Pro Gemini 3.1 Pro Thinking (High), No tools Vendor-reported | 94.3% |
| 5 | Claude Opus 4.7 Anthropic system card: 'averaged over 10 trials' Vendor-reported | 94.2% |
| 6 | Claude Mythos 5 Claude Fable 5 & Claude Mythos 5 System Card section 8.8: 'averaged over 5 trials'; Anthropic notes it considers GPQA Diamond saturated and plans to stop reporting it (same 94.1% shown for Claude Mythos 5 in OpenAI's GPT-5.6 table) Vendor-reported | 94.1% |
| 7 | GPT-5.5 'Evals of GPT were run with reasoning effort set to xhigh' in a research environment Vendor-reported | 93.6% |
| 8 | Claude Opus 4.8 Anthropic system card: 'averaged over 25 trials' Vendor-reported | 93.6% |
| 9 | GPT-5.6 Terra OpenAI GPT-5.6 launch eval table Vendor-reported | 92.9% |
| 10 | Qwen3.8-Max Qwen3.8-Max full benchmark table (weights published as Qwen/Qwen3.8-2.4T-A95B) Vendor-reported | 92.6 |
| 11 | DeepSeek-V4-Pro GPQA Diamond (Pass@1), 'Comparison with frontier models (Max reasoning effort)' table in DeepSeek-V4.1-Flash model card (original V4-Pro card lists DS-V4-Pro Max at 90.1) Vendor-reported | 92.4 |
| 12 | GPT-5.6 Luna OpenAI GPT-5.6 launch eval table Vendor-reported | 92.3% |
| 13 | Gemini 3 Pro Gemini 3 Pro Thinking (High), No tools (as listed in the Gemini 3.1 Pro model card and the Gemini 3 launch post) Vendor-reported | 91.9% |
| 14 | Qwen3.8-Flash-Next Qwen3.8-Flash-Next Language benchmark table (Scientific reasoning) Vendor-reported | 91.7 |
| 15 | GLM-5.2 Z.ai full benchmark table; reasoning tasks use temperature=1.0, top_p=0.95, max generation length 163,840 tokens; open weights Vendor-reported | 91.2 |
| 16 | DeepSeek-V4.1-Flash GPQA Diamond (Pass@1), Max reasoning effort; open weights Vendor-reported | 90.9 |
| 17 | Gemini 3 Flash Gemini 3 Flash launch post Vendor-reported | 90.4% |
| 18 | Qwen3.8-27B Qwen model card Text Performance table (Scientific reasoning); open weights Vendor-reported | 89.2 |
| 19 | Muse Spark Open (816B-A42B) Muse Spark Open (xhigh): 'xhigh reasoning effort, temperature = 1.0, top_p=1.0, top_k=disabled across benchmarks'; open weights Vendor-reported | 89.0 |
| 20 | Gemma 4 31B Gemma 4 model card benchmark results; open weights Vendor-reported | 84.3% |
About this board
- Rows marked vendor-reported come from the model maker's own model card or launch post, run under its own settings. Read the setting under each model before comparing two rows.
- No official GPQA Diamond leaderboard exists (the benchmark authors publish only the paper arXiv:2311.12022 and the github.com/idavidrein/gpqa repo); all entries are vendor-reported from model cards and release posts. Third-party aggregator leaderboards were excluded.
- Claude Opus 5.5, Claude Sonnet 5.5, Claude Haiku 5.5, Claude Fable 5.1 (self-reported): Anthropic launch-page benchmark grids omit GPQA Diamond, and the Fable 5 & Mythos 5 system card says Anthropic considers GPQA Diamond saturated and plans to stop reporting it. Fable 5.1, Opus 5 and Fable 5 values come from OpenAI comparison tables instead.
- GPT-6.1 Sol, GPT-6 Sol, GPT-6 Luna: their OpenAI launch posts have no GPQA Diamond row.
- Gemini 3.8 Flash and Gemini 3.7 Flash: Google's own model card (published 2 September 2026) does not list GPQA Diamond; the 95.3% figure is from OpenAI's GPT-6 Astra table.
- GLM-5.3: the zai-org/GLM-5.3 model card benchmark table has no GPQA Diamond row (Meta's Muse Spark Open card lists GLM 5.3 (max) at 91.7, DeepSeek's card lists 88.1; neither is Z.ai's own figure).
- MiniMax-M3: the official MiniMaxAI/MiniMax-M3 model card could not be read; only third-party figures were found (Z.ai lists 93.0).
- xAI Grok 4.5 / 4.6: no official xAI GPQA Diamond figure found; the latest on x.ai is Grok 4 at 87.5% (pass@1), which is not current.
- gpt-oss-120b / gpt-oss-20b: the GPQA Diamond values on the OpenAI gpt-oss post could not be extracted cleanly.
- Release dates could not be verified for GPT-6 Astra (post has later update notes dated September 22 and 29, 2026), Claude Mythos Preview, Claude Opus 4.7, Claude Fable 5.1 in OpenAI's table, and Gemma 4 model card; date left blank.
- Gemini 3 Pro (2025-11-18) and Gemini 3 Flash (2025-12-17) dates are launch dates as recalled, not re-read from the pages today; Muse Spark Open date (2026-10-15) came from a page summary and should be spot-checked.
FAQ
What does GPQA Diamond measure?
GPQA is a dataset of 448 multiple-choice questions written by domain experts in biology, physics, and chemistry, and GPQA Diamond is its 198-question subset where both expert validators answered correctly and at most one of three non-experts did.
Which model leads GPQA Diamond?
As of October 17, 2026, GPT-6 Astra is listed first on GPQA Diamond with a vendor-reported 96.0%. Vendors run the test under their own settings, so the list order is not a controlled comparison.
Where do these GPQA Diamond scores come from?
20 are vendor-reported, taken from the model maker's own model card or launch post. Every entry links to its source, which was read on October 17, 2026, and lists the setting it was run under.
Related reading
- LFM2.5-24B-A2B vs Qwen3.6-35B-A3B: Fast Local Agent or Better Coder?LFM2.5-24B-A2B vs Qwen3.6-35B-A3B for local agents: Liquid wins tool-use benchmarks, Qwen wins SWE-Bench Verified. Which one to run for which job.
- Claude Advisor Tool vs a Bigger Model: When the Pairing PaysAnthropic's own runs: a Fable 5.1 advisor added 1.7 points to Opus 5.5 for 2.1x the cost. When the Claude advisor tool pays, and when to raise effort instead.
- OpenAI Decisions API vs Jev vs Clef: Price, Speed, Which to PickOpenAI Decisions API costs $0.10 per million input tokens, 2.4x Jev. Pick Jev for cheap text decisions, Clef-flash for speed, Decisions API for images.
- Run Qwen3.8-Flash-Next on a Gaming PC With StrataRun Qwen3.8-Flash-Next locally with Strata on a 12 GB GPU: hardware needs, speeds by quant, the license catch, and when the hosted API is the better call.
- Kolibri Release Guide: Aleph Alpha's 78B Open ModelKolibri is an Apache 2.0 English-German MoE with 78B total and 3B active parameters and a 1M context. Benchmarks, the vLLM command, and who should run it.
- Cloudflare Clef Decision Models: Jev-Compatible, Benchmarked and PricedCloudflare Clef and Clef-flash are Jev-compatible decision models on Workers AI. Pricing, latency and benchmarks against Jev, plus the RL fine-tuning service.
Sources
Last verified . Each row links to the page its score was copied from.
- https://openai.com/index/gpt-6-astra/
- https://openai.com/index/gpt-5-6/
- https://www-cdn.anthropic.com/8b8380204f74670be75e81c820ca8dda846ab289.pdf
- https://deepmind.google/models/model-cards/gemini-3-1-pro/
- https://www-cdn.anthropic.com/037f06850df7fbe871e206dad004c3db5fd50340/Claude%20Opus%204.7%20System%20Card.pdf
- https://www-cdn.anthropic.com/d00db56fa754a1b115b6dd7cb2e3c342ee809620.pdf
- https://openai.com/index/introducing-gpt-5-5/
- https://www-cdn.anthropic.com/0b4915911bb0d19eca5b5ee635c80fef830a37ea.pdf
- https://qwen.ai/blog?id=qwen3.8
- https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash
- https://blog.google/products-and-platforms/products/gemini/gemini-3/
- https://qwen.ai/blog?id=qwen3.8-flash-next
- https://z.ai/blog/glm-5.2
- https://blog.google/products-and-platforms/products/gemini/gemini-3-flash/
- https://huggingface.co/Qwen/Qwen3.8-27B
- https://huggingface.co/meta-models/Muse-Spark-Open-816B-A42B
- https://ai.google.dev/gemma/docs/core/model_card_4
Newsletter
Get new benchmark results as they land
One email per week. New model releases, the leaderboard moves that matter and what they mean for what you build.