Calibration
Gert Labs → Agents' Last Exam
Agents' Last Exam is estimated from Gert Labs with a Hill curve fitted on 16 models measured on both: y = 0.2956 + (0.8314 − 0.2956)·x^6.00 / (0.86661^6.00 + x^6.00), R² = 0.58, cross-validated error 9.7 pp. It is used for 90 estimates.
| Estimated model | Gert Labs | Agents' Last Exam | Source |
|---|---|---|---|
| GPT-6.2 Astra | 90.0% | 59.4% | estimated ± 9.7 pp, medium confidence |
| Claude Fable 5.2 | 88.2% | 57.8% | estimated ± 9.7 pp, medium confidence |
| GPT-6.2 Sol | 86.8% | 56.5% | estimated ± 9.7 pp, medium confidence |
| Gemini 4 Neon | 83.8% | 53.7% | estimated ± 9.7 pp, medium confidence |
| Gemini 4 Hydrogen | 79.1% | 49.2% | estimated ± 9.7 pp, medium confidence |
| Claude Opus 5 | 77.4% | 47.6% | estimated ± 9.7 pp, medium confidence |
| Claude Fable 5.1 | 77.0% | 47.3% | estimated ± 9.7 pp, medium confidence |
| Claude Fable 5 | 72.7% | 43.4% | estimated ± 9.7 pp, medium confidence |
| Grok 4.7 | 71.5% | 42.4% | estimated ± 9.7 pp, medium confidence |
| Qwen4 Max | 70.2% | 41.3% | estimated ± 9.7 pp, medium confidence |
| Claude Sonnet 5.5 | 69.9% | 41.1% | estimated ± 9.7 pp, medium confidence |
| Grok 4.6 | 69.2% | 40.6% | estimated ± 9.7 pp, medium confidence |
| Gemini 3.8 Flash | 67.0% | 39.0% | estimated ± 9.7 pp, medium confidence |
| Muse Wonder | 65.9% | 38.3% | estimated ± 9.7 pp, medium confidence |
| Kimi K3 | 63.1% | 36.5% | estimated ± 9.7 pp, medium confidence |
| Gemini 3.6 Flash | 59.5% | 34.6% | estimated ± 9.7 pp, medium confidence |
| Claude Haiku 5.5 | 58.6% | 34.2% | estimated ± 9.7 pp, medium confidence |
| Ember-1 | 58.0% | 34.0% | estimated ± 9.7 pp, medium confidence |
| Muse Spark 1.3 | 57.0% | 33.6% | estimated ± 9.7 pp, medium confidence |
| Muse Ember | 56.1% | 33.2% | estimated ± 9.7 pp, medium confidence |
| Claude Opus 4.8 | 55.6% | 33.0% | estimated ± 9.7 pp, medium confidence |
| GPT-5.3 Codex | 52.0% | 31.9% | estimated ± 9.7 pp, medium confidence |
| Mercury 3 | 50.6% | 31.6% | estimated ± 9.7 pp, medium confidence |
| GLM-5.2 | 49.1% | 31.3% | estimated ± 9.7 pp, medium confidence |
| Inkling 2 Small | 48.6% | 31.2% | estimated ± 9.7 pp, medium confidence |
| Gemini 3.5 Flash | 48.6% | 31.2% | estimated ± 9.7 pp, medium confidence |
| GPT-5.2 | 47.7% | 31.0% | estimated ± 9.7 pp, medium confidence |
| Qwen 3.6 Max (preview) | 47.5% | 31.0% | estimated ± 9.7 pp, medium confidence |
| Gemini 3 Pro | 47.4% | 31.0% | estimated ± 9.7 pp, medium confidence |
| Grok 4.5 | 47.2% | 30.9% | estimated ± 9.7 pp, medium confidence |
| Claude Sonnet 5 | 47.0% | 30.9% | estimated ± 9.7 pp, medium confidence |
| Kimi K2.6 | 47.0% | 30.9% | estimated ± 9.7 pp, medium confidence |
| Claude Opus 4.5 | 46.0% | 30.7% | estimated ± 9.7 pp, medium confidence |
| Gemini 3.1 Pro | 45.7% | 30.7% | estimated ± 9.7 pp, medium confidence |
| Muse Spark 1.2 | 45.2% | 30.6% | estimated ± 9.7 pp, medium confidence |
| Qwen3.7 Max | 44.8% | 30.6% | estimated ± 9.7 pp, medium confidence |
| MiMo-V2.5-Pro | 44.2% | 30.5% | estimated ± 9.7 pp, medium confidence |
| Muse Spark Open | 43.6% | 30.4% | estimated ± 9.7 pp, medium confidence |
| Kimi K2.7 Code | 40.7% | 30.1% | estimated ± 9.7 pp, medium confidence |
| GPT-5.1 | 39.9% | 30.1% | estimated ± 9.7 pp, medium confidence |
| Mistral Large 4 | 39.8% | 30.1% | estimated ± 9.7 pp, medium confidence |
| Hy3 Preview | 39.3% | 30.0% | estimated ± 9.7 pp, medium confidence |
| Beam 1.0 501B-A23B | 36.9% | 29.9% | estimated ± 9.7 pp, medium confidence |
| Kimi K2.5 | 35.5% | 29.8% | estimated ± 9.7 pp, medium confidence |
| Qwen3.7 Plus | 35.3% | 29.8% | estimated ± 9.7 pp, medium confidence |
| Claude Sonnet 4.5 | 34.9% | 29.8% | estimated ± 9.7 pp, medium confidence |
| Gemini 3 Flash | 34.2% | 29.8% | estimated ± 9.7 pp, low confidence |
| Muse Glimmer 30B | 32.7% | 29.7% | estimated ± 9.7 pp, low confidence |
| Gemini 2.5 Pro | 32.6% | 29.7% | estimated ± 9.7 pp, low confidence |
| Qwen3.6-27B | 32.4% | 29.7% | estimated ± 9.7 pp, low confidence |
| Grok 4 | 31.6% | 29.7% | estimated ± 9.7 pp, low confidence |
| Qwen3.6 Plus | 31.4% | 29.7% | estimated ± 9.7 pp, low confidence |
| Nemotron 3.5 Super VL 120B A12B | 30.9% | 29.7% | estimated ± 9.7 pp, low confidence |
| Grok 4.20 Multi-agent | 30.2% | 29.7% | estimated ± 9.7 pp, low confidence |
| MiMo-V2.5 | 29.9% | 29.7% | estimated ± 9.7 pp, low confidence |
| MiniMax M3 | 29.8% | 29.6% | estimated ± 9.7 pp, low confidence |
| Grok 4.1 Fast | 29.3% | 29.6% | estimated ± 9.7 pp, low confidence |
| Qwen3.5 397B | 29.2% | 29.6% | estimated ± 9.7 pp, low confidence |
| Qwen3.5-27B | 28.9% | 29.6% | estimated ± 9.7 pp, low confidence |
| MiniMax M2.7 | 28.8% | 29.6% | estimated ± 9.7 pp, low confidence |
| GPT-5.1-Codex | 28.1% | 29.6% | estimated ± 9.7 pp, low confidence |
| Claude 4 Sonnet | 27.7% | 29.6% | estimated ± 9.7 pp, low confidence |
| Gemma 4 31B | 27.7% | 29.6% | estimated ± 9.7 pp, low confidence |
| GLM-5 | 27.7% | 29.6% | estimated ± 9.7 pp, low confidence |
| GPT-5.2-Codex | 27.3% | 29.6% | estimated ± 9.7 pp, low confidence |
| Nemotron 3 Ultra | 27.3% | 29.6% | estimated ± 9.7 pp, low confidence |
| Grok Build 0.1 | 27.2% | 29.6% | estimated ± 9.7 pp, low confidence |
| Inkling | 26.6% | 29.6% | estimated ± 9.7 pp, low confidence |
| Hy3 | 26.3% | 29.6% | estimated ± 9.7 pp, low confidence |
| Gemini 3.1 Flash-Lite | 25.7% | 29.6% | estimated ± 9.7 pp, low confidence |
| Trinity-Large-Thinking | 25.4% | 29.6% | estimated ± 9.7 pp, low confidence |
| MiMo-V2-Pro | 24.7% | 29.6% | estimated ± 9.7 pp, low confidence |
| Qwen3.7 Flash | 24.6% | 29.6% | estimated ± 9.7 pp, low confidence |
| GLM-4.7 | 23.8% | 29.6% | estimated ± 9.7 pp, low confidence |
| Gemini 3.5 Flash-Lite | 23.8% | 29.6% | estimated ± 9.7 pp, low confidence |
| GLM-5V-Turbo | 23.5% | 29.6% | estimated ± 9.7 pp, low confidence |
| Grok 4.20 | 22.3% | 29.6% | estimated ± 9.7 pp, low confidence |
| Qwen3.6-35B-A3B | 21.5% | 29.6% | estimated ± 9.7 pp, low confidence |
| DeepSeek V3.2 | 21.1% | 29.6% | estimated ± 9.7 pp, low confidence |
| GPT-OSS 120B | 21.1% | 29.6% | estimated ± 9.7 pp, low confidence |
| Grok 4.3 | 21.0% | 29.6% | estimated ± 9.7 pp, low confidence |
| Mistral Medium 3.5 128B | 20.7% | 29.6% | estimated ± 9.7 pp, low confidence |
| Laguna M.1 | 20.3% | 29.6% | estimated ± 9.7 pp, low confidence |
| Step 3.7 Flash | 20.3% | 29.6% | estimated ± 9.7 pp, low confidence |
| Qwen3.5-35B-A3B | 20.2% | 29.6% | estimated ± 9.7 pp, low confidence |
| Mercury 2.5 | 20.0% | 29.6% | estimated ± 9.7 pp, low confidence |
| Gemma 4 26B A4B | 19.6% | 29.6% | estimated ± 9.7 pp, low confidence |
| GPT-4.1 | 18.9% | 29.6% | estimated ± 9.7 pp, low confidence |
| Laguna S 2.1 | 17.5% | 29.6% | estimated ± 9.7 pp, low confidence |
| Laguna XS 2.1 | 17.0% | 29.6% | estimated ± 9.7 pp, low confidence |
Fitted on
| Model | Gert Labs | Agents' Last Exam |
|---|---|---|
| DeepSeek V4.1 Flash 1017 | 60.1% | 35.4% |
| DeepSeek V4 Flash 0731 | 46.8% | 25.2% |
| DeepSeek V4 Pro | 52.7% | 25.7% |
| Gemini 3.7 Flash | 61.4% | 26.3% |
| Gemini 4 Argon | 80.6% | 39.5% |
| GLM-5.3 | 55.0% | 28.5% |
| GLM-5.3-Flash | 49.6% | 26.3% |
| GPT-6.1 Terra | 83.2% | 60.0% |
| GPT-6 Astra | 92.0% | 59.3% |
| GPT-6 Sol | 79.0% | 56.4% |
| Kimi K3.1 Pro | 65.8% | 43.2% |
| MiMo-V2.6-Flash | 53.5% | 27.6% |
| MiMo-V2.6-Pro | 59.5% | 31.6% |
| MiniMax M3.1 Flash | 42.5% | 29.6% |
| Qwen3.8-27B | 34.5% | 42.9% |
| Qwen3.8 Max | 55.7% | 52.4% |