Benchmarks · knowledge
GPQA Diamond
GPQA Diamond: graduate-level science multiple choice; accuracy.
131 models · 224 results · data as of 2026-10-29 · artificialanalysis.ai ↗
Leaderboard
GPT-6 Astra xhigh
More from this benchmark
3 published measuresScore vs any published measure3 measures: sub-scores, splits, counts
Breakdown
darker is better within each column · click a column to sort| Model | |||
|---|---|---|---|
| GPT-6 Astra xhigh | 49,697 | 32,146 | 245,259 |
| Gemini 3.8 Flash high | 49,283 | 95,304 | 1,192,128 |
| Grok 4.6 high | 53,480 | 44,652 | 1,137,270 |
| Gemini 3.7 Flash high | 49,283 | 84,947 | 431,277 |
| GPT-5.6 Sol max | 63,940 | 25,036 | 686,387 |
| Muse Spark 1.3 xhigh | 47,342 | 47,499 | 1,325,792 |
| Fable 5.1 max, fallback | 71,502 | 89,802 | 657,956 |
| Opus 5 high | 71,195 | 59,723 | 284,261 |
| GPT-5.5 xhigh | 52,903 | 34,358 | 973,392 |
| Kimi K3 max | 65,512 | 33,027 | 1,950,185 |
| Qwen 3.8 2.4T-A95B | 61,974 | 45,364 | 2,105,723 |
| Grok 4.5 high | 53,480 | 64,141 | 959,467 |
| MiniMax M3 | 80,656 | 90,942 | 1,520,405 |
| Gemini 3.6 Flash high | 49,283 | 99,703 | 683,650 |
| GPT-5.6 Terra max | 145,520 | 24,945 | 1,154,803 |
| Qwen 3.7 Max | 51,876 | 109,278 | 2,034,811 |
| Gemini 3.5 Flash high | 49,283 | 111,337 | 986,006 |
| Opus 4.8 max | 70,113 | 135,763 | 2,323,532 |
| GPT-5.4 xhigh | 49,132 | 41,903 | 1,160,668 |
| GLM 5.3 max | 51,246 | 29,939 | 1,376,584 |
| GPT-5.3 Codex xhigh | 49,132 | 36,946 | 1,157,885 |
| Opus 4.7 max | 76,991 | 88,573 | 2,494,337 |
| GLM 5.3 Flash max | 51,246 | 30,319 | 2,132,857 |
| Sonnet 5 max | 70,843 | 180,641 | 7,057,185 |
| GPT-5.6 Luna max | 336,251 | 24,143 | 1,554,866 |
All results
224 rows| # | Model | Effort | GPQA Diamond |
|---|---|---|---|
| 1 | GPT-6 Astra | xhigh | 96.3% |
| 2 | GPT-6 Astra | max | 96.1% |
| 3 | Gemini 3.8 Flash | high | 95.3% |
| 4 | GPT-6 Astra | high | 94.9% |
| 5 | Grok 4.6 | high | 94.9% |
| 6 | Gemini 3.7 Flash | high | 94.5% |
| 7 | GPT-5.6 Sol | max | 94.1% |
| 8 | Muse Spark 1.3 | xhigh | 94.1% |
| 9 | GPT-6 Astra | medium | 93.9% |
| 10 | Fable 5.1 | max | 93.7% |
| 11 | Opus 5 | high | 93.7% |
| 12 | Opus 5 | xhigh | 93.7% |
| 13 | Gemini 3.8 Flash | medium | 93.5% |
| 14 | GPT-5.5 | xhigh | 93.5% |
| 15 | Grok 4.6 | medium | 93.5% |
| 16 | Grok 4.6 | xhigh | 93.5% |
| 17 | Kimi K3 | max | 93.5% |
| 18 | Muse Spark 1.3 | max | 93.5% |
| 19 | Qwen 3.8 2.4T-A95B | 93.5% | |
| 20 | Fable 5.1 | xhigh | 93.4% |
| 21 | Opus 5 | max | 93.2% |
| 22 | GPT-5.5 | high | 93.2% |
| 23 | GPT-5.6 Sol | xhigh | 93.1% |
| 24 | GPT-6 Astra | low | 93.1% |
| 25 | Grok 4.5 | high | 93.1% |
| 26 | MiniMax M3 | 92.9% | |
| 27 | Gemini 3.6 Flash | high | 92.8% |
| 28 | GPT-5.6 Sol | high | 92.8% |
| 29 | GPT-5.5 | medium | 92.6% |
| 30 | GPT-5.6 Sol | medium | 92.6% |