# GPQA → CritPt

Calibration · https://benchgap.net/calibration/gpqa/current/to/critpt/current · data retrieved 2026-11-18

CritPt is estimated from GPQA with a offset logistic curve fitted on 84 models measured on both: y = 0.0033 + (0.4421 − 0.0033) / (1 + exp(−40.35·(x − 0.9360))), R² = 0.82, cross-validated error 4.0 pp. It is used for 42 estimates.

| Estimated model | GPQA | CritPt | Source |
|---|---|---|---|
| [GPT-6.2 Astra](https://benchgap.net/model/gpt-6-2-astra.md) | 97.0% | 35.3% | estimated ± 4.0 pp, medium confidence |
| [GPT-6.2 Sol](https://benchgap.net/model/gpt-6-2-sol.md) | 96.0% | 32.1% | estimated ± 4.0 pp, high confidence |
| [Sakana Fugu](https://benchgap.net/model/sakana-fugu.md) | 95.5% | 30.3% | estimated ± 4.0 pp, high confidence |
| [Sakana Fugu-Ultra](https://benchgap.net/model/sakana-fugu-ultra.md) | 95.5% | 30.3% | estimated ± 4.0 pp, high confidence |
| [Claude Mythos 5](https://benchgap.net/model/claude-mythos-5.md) | 94.1% | 24.5% | estimated ± 4.0 pp, high confidence |
| [Ornith-1.5-397B](https://benchgap.net/model/ornith-1-5-397b.md) | 92.8% | 18.8% | estimated ± 4.0 pp, high confidence |
| [Qwen3.8 Max](https://benchgap.net/model/qwen3-8-max.md) | 92.6% | 17.9% | estimated ± 4.0 pp, high confidence |
| [Qwen3.8-Omni-Flash](https://benchgap.net/model/qwen3-8-omni-flash.md) | 91.0% | 11.7% | estimated ± 4.0 pp, high confidence |
| [Ornith-1.5-35B-A3B](https://benchgap.net/model/ornith-1-5-35b-a3b.md) | 89.2% | 6.7% | estimated ± 4.0 pp, high confidence |
| [Ornith-1.5-9B](https://benchgap.net/model/ornith-1-5-9b.md) | 86.4% | 2.6% | estimated ± 4.0 pp, high confidence |
| [Ternary Bonsai 2 27B](https://benchgap.net/model/ternary-bonsai-2-27b.md) | 85.8% | 2.1% | estimated ± 4.0 pp, high confidence |
| [MAI-Thinking-1](https://benchgap.net/model/mai-thinking-1.md) | 84.2% | 1.3% | estimated ± 4.0 pp, high confidence |
| [Ling 3.0 Flash FP8](https://benchgap.net/model/ling-3-0-flash-fp8.md) | 84.0% | 1.2% | estimated ± 4.0 pp, high confidence |
| [Claude Sonnet 4.5](https://benchgap.net/model/claude-sonnet-4-5.md) | 83.4% | 1.0% | estimated ± 4.0 pp, high confidence |
| [DeepSeek-R1-0528](https://benchgap.net/model/deepseek-r1-0528.md) | 81.0% | 0.6% | estimated ± 4.0 pp, high confidence |
| [Grok 3 Mini](https://benchgap.net/model/grok-3-mini.md) | 80.3% | 0.5% | estimated ± 4.0 pp, high confidence |
| [Nemotron 3 Super 100B](https://benchgap.net/model/nemotron-3-super-100b.md) | 79.2% | 0.5% | estimated ± 4.0 pp, high confidence |
| [o1-pro](https://benchgap.net/model/o1-pro.md) | 79.0% | 0.5% | estimated ± 4.0 pp, high confidence |
| [Qwen3 235B 2507](https://benchgap.net/model/qwen3-235b-2507.md) | 77.5% | 0.4% | estimated ± 4.0 pp, high confidence |
| [Grok 3 [Beta]](https://benchgap.net/model/grok-3-beta.md) | 75.4% | 0.4% | estimated ± 4.0 pp, high confidence |
| [GLM-4.7-Flash](https://benchgap.net/model/glm-4-7-flash.md) | 75.2% | 0.4% | estimated ± 4.0 pp, high confidence |
| [LFM2.5-24B-A2B](https://benchgap.net/model/lfm2-5-24b-a2b.md) | 74.4% | 0.3% | estimated ± 4.0 pp, high confidence |
| [ZAYA1-8B](https://benchgap.net/model/zaya1-8b.md) | 71.0% | 0.3% | estimated ± 4.0 pp, high confidence |
| [Spark-X2.6-4B](https://benchgap.net/model/spark-x2-6-4b.md) | 69.9% | 0.3% | estimated ± 4.0 pp, high confidence |
| [GPT-4.5](https://benchgap.net/model/gpt-4-5.md) | 69.5% | 0.3% | estimated ± 4.0 pp, high confidence |
| [Phi-4-reasoning-plus](https://benchgap.net/model/phi-4-reasoning-plus.md) | 68.9% | 0.3% | estimated ± 4.0 pp, high confidence |
| [LFM3-1.7B](https://benchgap.net/model/lfm3-1-7b.md) | 63.4% | 0.3% | estimated ± 4.0 pp, high confidence |
| [Qwen2.5-Max](https://benchgap.net/model/qwen2-5-max.md) | 60.1% | 0.3% | estimated ± 4.0 pp, high confidence |
| [Claude 3.5 Sonnet](https://benchgap.net/model/claude-3-5-sonnet.md) | 59.4% | 0.3% | estimated ± 4.0 pp, high confidence |
| [Mellum2-12B-A2.5B-Thinking](https://benchgap.net/model/mellum2-12b-a2-5b-thinking.md) | 57.6% | 0.3% | estimated ± 4.0 pp, high confidence |
| [ZAYA1-74B-Preview](https://benchgap.net/model/zaya1-74b-preview.md) | 57.3% | 0.3% | estimated ± 4.0 pp, high confidence |
| [Grok 2](https://benchgap.net/model/grok-2.md) | 56.0% | 0.3% | estimated ± 4.0 pp, high confidence |
| [Grok 2 mini](https://benchgap.net/model/grok-2-mini.md) | 51.0% | 0.3% | estimated ± 4.0 pp, high confidence |
| [Gemini 2.0 Flash-Lite](https://benchgap.net/model/gemini-2-0-flash-lite.md) | 50.5% | 0.3% | estimated ± 4.0 pp, high confidence |
| [Qwen2.5-VL-32B](https://benchgap.net/model/qwen2-5-vl-32b.md) | 46.0% | 0.3% | estimated ± 4.0 pp, high confidence |
| [Soofi S 30B-A3B](https://benchgap.net/model/soofi-s-30b-a3b.md) | 43.4% | 0.3% | estimated ± 4.0 pp, high confidence |
| [Mellum2-12B-A2.5B-Instruct](https://benchgap.net/model/mellum2-12b-a2-5b-instruct.md) | 40.9% | 0.3% | estimated ± 4.0 pp, high confidence |
| [LFM2.5-350M](https://benchgap.net/model/lfm2-5-350m.md) | 30.6% | 0.3% | estimated ± 4.0 pp, high confidence |
| [Phi-3.5-mini](https://benchgap.net/model/phi-3-5-mini.md) | 30.4% | 0.3% | estimated ± 4.0 pp, high confidence |
| [Granite-4.0-1B](https://benchgap.net/model/granite-4-0-1b.md) | 29.9% | 0.3% | estimated ± 4.0 pp, high confidence |
| [LFM2.5-VL-450M](https://benchgap.net/model/lfm2-5-vl-450m.md) | 25.7% | 0.3% | estimated ± 4.0 pp, high confidence |
| [LFM2.5-230M](https://benchgap.net/model/lfm2-5-230m.md) | 25.4% | 0.3% | estimated ± 4.0 pp, high confidence |

## Fitted on

| Model | GPQA | CritPt |
|---|---|---|
| [Claude 3.5 Haiku](https://benchgap.net/model/claude-3-5-haiku.md) | 41.6% | 0.0% |
| [Claude 3.7 Sonnet](https://benchgap.net/model/claude-3-7-sonnet.md) | 68.0% | 0.0% |
| [Claude 3.7 Sonnet Thinking](https://benchgap.net/model/claude-3-7-sonnet-thinking.md) | 78.2% | 0.9% |
| [Claude Opus 4.5](https://benchgap.net/model/claude-opus-4-5.md) | 87.0% | 0.3% |
| [Claude Opus 4.6](https://benchgap.net/model/claude-opus-4-6.md) | 91.3% | 2.8% |
| [Claude Opus 4.7](https://benchgap.net/model/claude-opus-4-7.md) | 94.2% | 12.0% |
| [Claude Opus 4.8](https://benchgap.net/model/claude-opus-4-8.md) | 93.6% | 20.9% |
| [Claude Sonnet 4.6](https://benchgap.net/model/claude-sonnet-4-6.md) | 89.9% | 0.9% |
| [DeepSeek-R1](https://benchgap.net/model/deepseek-r1.md) | 71.5% | 1.4% |
| [DeepSeek V3](https://benchgap.net/model/deepseek-v3.md) | 59.1% | 0.0% |
| [DeepSeek V3.1](https://benchgap.net/model/deepseek-v3-1.md) | 74.9% | 0.0% |
| [DeepSeek V4.1 Flash 1017](https://benchgap.net/model/deepseek-v4-1-flash.md) | 90.9% | 14.3% |
| [DeepSeek V4 Flash 0731](https://benchgap.net/model/deepseek-v4-flash-0731.md) | 88.1% | 16.6% |
| [DeepSeek V4 Pro](https://benchgap.net/model/deepseek-v4-pro.md) | 90.1% | 18.0% |
| [Gemini 2.0 Flash](https://benchgap.net/model/gemini-2-0-flash.md) | 65.2% | 0.0% |
| [Gemini 2.5 Flash-Lite](https://benchgap.net/model/gemini-2-5-flash-lite.md) | 64.6% | 0.0% |
| [Gemini 2.5 Pro](https://benchgap.net/model/gemini-2-5-pro.md) | 83.0% | 2.6% |
| [Gemini 3.5 Flash](https://benchgap.net/model/gemini-3-5-flash.md) | 92.2% | 13.1% |
| [Gemma 4 12B](https://benchgap.net/model/gemma-4-12b.md) | 78.8% | 0.0% |
| [Gemma 4 26B A4B](https://benchgap.net/model/gemma-4-26b-a4b.md) | 82.3% | 0.0% |
| [Gemma 4 31B](https://benchgap.net/model/gemma-4-31b.md) | 84.3% | 1.4% |
| [Gemma 4 E2B](https://benchgap.net/model/gemma-4-e2b.md) | 43.4% | 0.0% |
| [Gemma 4 E4B](https://benchgap.net/model/gemma-4-e4b.md) | 58.6% | 0.6% |
| [GLM-4.7](https://benchgap.net/model/glm-4-7.md) | 85.7% | 1.7% |
| [GLM-5](https://benchgap.net/model/glm-5.md) | 86.0% | 2.0% |
| [GLM-5.2](https://benchgap.net/model/glm-5-2.md) | 91.2% | 20.9% |
| [GPT-4.1](https://benchgap.net/model/gpt-4-1.md) | 66.3% | 0.0% |
| [GPT-4.1 mini](https://benchgap.net/model/gpt-4-1-mini.md) | 64.2% | 0.0% |
| [GPT-4.1 nano](https://benchgap.net/model/gpt-4-1-nano.md) | 50.3% | 0.0% |
| [GPT-5.2](https://benchgap.net/model/gpt-5-2.md) | 92.4% | 11.6% |
| [GPT-5.4](https://benchgap.net/model/gpt-5-4.md) | 92.8% | 23.4% |
| [GPT-5.4 mini](https://benchgap.net/model/gpt-5-4-mini.md) | 88.0% | 10.0% |
| [GPT-5.4 nano](https://benchgap.net/model/gpt-5-4-nano.md) | 82.8% | 9.3% |
| [GPT-5.5](https://benchgap.net/model/gpt-5-5.md) | 93.6% | 27.1% |
| [GPT-5.6 Luna](https://benchgap.net/model/gpt-5-6-luna.md) | 92.3% | 20.6% |
| [GPT-5.6 Sol](https://benchgap.net/model/gpt-5-6-sol.md) | 94.6% | 32.3% |
| [GPT-5.6 Terra](https://benchgap.net/model/gpt-5-6-terra.md) | 92.9% | 30.0% |
| [GPT-6 Astra](https://benchgap.net/model/gpt-6-astra.md) | 96.0% | 31.7% |
| [GPT-OSS 120B](https://benchgap.net/model/gpt-oss-120b.md) | 80.1% | 1.1% |
| [GPT-OSS 20B](https://benchgap.net/model/gpt-oss-20b.md) | 71.5% | 1.4% |
| [Granite-4.0-350M](https://benchgap.net/model/granite-4-0-350m.md) | 24.1% | 0.0% |
| [Granite-4.0-H-1B](https://benchgap.net/model/granite-4-0-h-1b.md) | 29.7% | 0.0% |
| [Granite-4.0-H-350M](https://benchgap.net/model/granite-4-0-h-350m.md) | 26.1% | 0.0% |
| [Granite 4.2 30B](https://benchgap.net/model/granite-4-2-30b.md) | 66.4% | 0.3% |
| [Granite 4.2 3B](https://benchgap.net/model/granite-4-2-3b.md) | 54.8% | 0.0% |
| [Granite 4.2 8B](https://benchgap.net/model/granite-4-2-8b.md) | 64.1% | 0.3% |
| [Grok 4.3](https://benchgap.net/model/grok-4-3.md) | 90.1% | 8.0% |
| [Hy3 Preview](https://benchgap.net/model/hy3-preview.md) | 87.2% | 4.9% |
| [Hy4 preview](https://benchgap.net/model/hy4-preview.md) | 92.3% | 16.9% |
| [Inkling](https://benchgap.net/model/inkling.md) | 87.9% | 5.4% |
| [Inkling-Small](https://benchgap.net/model/inkling-small.md) | 89.5% | 8.3% |
| [Kimi K2.6](https://benchgap.net/model/kimi-2-6.md) | 90.5% | 8.0% |
| [Kimi K3.1 Pro](https://benchgap.net/model/kimi-3-1-pro.md) | 94.2% | 24.6% |
| [Kimi K2](https://benchgap.net/model/kimi-k2.md) | 75.1% | 0.0% |
| [Kimi K2.5](https://benchgap.net/model/kimi-k2-5.md) | 87.6% | 3.1% |
| [Kimi K3](https://benchgap.net/model/kimi-k3.md) | 93.5% | 23.4% |
| [Ling 2.6 Flash](https://benchgap.net/model/ling-2-6-flash.md) | 59.0% | 0.0% |
| [Ling 3.0 Flash](https://benchgap.net/model/ling-3-0-flash.md) | 85.0% | 1.7% |
| [Llama 3.1 70B](https://benchgap.net/model/llama-3-1-70b.md) | 46.7% | 0.0% |
| [Llama 3.3 70B](https://benchgap.net/model/llama-3-3-70b.md) | 50.5% | 0.0% |
| [Magistral Medium](https://benchgap.net/model/magistral-medium.md) | 70.8% | 0.3% |
| [Magistral Small](https://benchgap.net/model/magistral-small.md) | 68.2% | 0.0% |
| [MiMo-V2-Flash](https://benchgap.net/model/mimo-v2-flash.md) | 83.7% | 0.0% |
| [Mistral Small 3.1](https://benchgap.net/model/mistral-small-3-1.md) | 46.0% | 0.0% |
| [Nemotron 3.5 Lightning 30B A3B NVFP4](https://benchgap.net/model/nemotron-3-5-lightning-30b-a3b-nvfp4.md) | 75.6% | 0.0% |
| [Nemotron 3 Nano 30B](https://benchgap.net/model/nemotron-3-nano-30b.md) | 73.0% | 0.0% |
| [Nemotron 3 Nano Omni 30B A3B](https://benchgap.net/model/nemotron-3-nano-omni-30b-a3b.md) | 72.2% | 0.0% |
| [Nemotron 3 Ultra](https://benchgap.net/model/nemotron-3-ultra.md) | 87.0% | 3.1% |
| [Nemotron Ultra 253B](https://benchgap.net/model/nemotron-ultra-253b.md) | 76.0% | 0.0% |
| [o1](https://benchgap.net/model/o1.md) | 75.7% | 0.3% |
| [o4-mini](https://benchgap.net/model/o4-mini.md) | 77.6% | 0.6% |
| [Phi-4](https://benchgap.net/model/phi-4.md) | 56.1% | 0.0% |
| [Phi-4-reasoning](https://benchgap.net/model/phi-4-reasoning.md) | 65.8% | 0.0% |
| [Qwen3.5-122B-A10B](https://benchgap.net/model/qwen3-5-122b-a10b.md) | 86.6% | 0.6% |
| [Qwen3.5-27B](https://benchgap.net/model/qwen3-5-27b.md) | 85.5% | 0.9% |
| [Qwen3.5-35B-A3B](https://benchgap.net/model/qwen3-5-35b-a3b.md) | 84.2% | 0.9% |
| [Qwen3.5 397B](https://benchgap.net/model/qwen3-5-397b.md) | 88.4% | 1.7% |
| [Qwen3.6-27B](https://benchgap.net/model/qwen3-6-27b.md) | 87.8% | 1.1% |
| [Qwen3.6-35B-A3B](https://benchgap.net/model/qwen3-6-35b-a3b.md) | 86.0% | 0.3% |
| [Qwen3.6 Plus](https://benchgap.net/model/qwen3-6-plus.md) | 90.4% | 2.9% |
| [Qwen3.7 Max](https://benchgap.net/model/qwen3-7-max.md) | 92.4% | 13.4% |
| [Qwen3.7 Plus](https://benchgap.net/model/qwen3-7-plus.md) | 90.3% | 9.1% |
| [Qwen3.8-27B](https://benchgap.net/model/qwen3-8-27b.md) | 89.2% | 5.4% |
| [Qwen3.8-Flash-Next](https://benchgap.net/model/qwen3-8-flash-next.md) | 91.7% | 11.1% |
