# BrowseComp → CyberGym

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

CyberGym is estimated from BrowseComp with a inverse Michaelis–Menten curve fitted on 15 models measured on both: y = 1.07522·(x − 0.0000) / (0.0000 + 2.0000 − x), R² = 0.60, cross-validated error 7.3 pp. It is used for 19 estimates.

| Estimated model | BrowseComp | CyberGym | Source |
|---|---|---|---|
| [GPT-5.5 Pro](https://benchgap.net/model/gpt-5-5-pro.md) | 90.1% | 88.2% | estimated ± 7.3 pp, medium confidence |
| [GPT-5.4 Pro](https://benchgap.net/model/gpt-5-4-pro.md) | 89.3% | 86.7% | estimated ± 7.3 pp, medium confidence |
| [dots3-note Preview](https://benchgap.net/model/dots3-note-preview.md) | 83.3% | 76.7% | estimated ± 7.3 pp, medium confidence |
| [Sage-A1.5](https://benchgap.net/model/sage-a1-5.md) | 81.7% | 74.3% | estimated ± 7.3 pp, medium confidence |
| [Ling 3.0 Flash Sante](https://benchgap.net/model/ling-3-0-flash-sante.md) | 80.2% | 72.0% | estimated ± 7.3 pp, medium confidence |
| [Beam 1.0 501B-A23B FP8](https://benchgap.net/model/reflection-beam-fp8.md) | 76.7% | 66.9% | estimated ± 7.3 pp, medium confidence |
| [Beam 1.0 501B-A23B NVFP4](https://benchgap.net/model/reflection-beam-nvfp4.md) | 75.6% | 65.3% | estimated ± 7.3 pp, medium confidence |
| [Agents-A1](https://benchgap.net/model/agents-a1.md) | 75.5% | 65.2% | estimated ± 7.3 pp, medium confidence |
| [Nemotron 3.5 Bolt 74B-A9B BF16](https://benchgap.net/model/nemotron-3-5-bolt-74b-a9b-bf16.md) | 70.0% | 57.9% | estimated ± 7.3 pp, medium confidence |
| [Qwen3.5 397B](https://benchgap.net/model/qwen3-5-397b.md) | 69.0% | 56.6% | estimated ± 7.3 pp, medium confidence |
| [Ornith-1.5-35B-A3B](https://benchgap.net/model/ornith-1-5-35b-a3b.md) | 67.6% | 54.9% | estimated ± 7.3 pp, medium confidence |
| [Agents-A1-4B](https://benchgap.net/model/agents-a1-4b.md) | 66.8% | 53.9% | estimated ± 7.3 pp, medium confidence |
| [GPT-5.2](https://benchgap.net/model/gpt-5-2.md) | 65.8% | 52.7% | estimated ± 7.3 pp, medium confidence |
| [Qwen3.5-35B-A3B](https://benchgap.net/model/qwen3-5-35b-a3b.md) | 61.0% | 47.2% | estimated ± 7.3 pp, low confidence |
| [Ornith-1.5-9B](https://benchgap.net/model/ornith-1-5-9b.md) | 56.4% | 42.2% | estimated ± 7.3 pp, low confidence |
| [Spark-X2.6-4B](https://benchgap.net/model/spark-x2-6-4b.md) | 53.1% | 38.9% | estimated ± 7.3 pp, low confidence |
| [Sarvam 105B](https://benchgap.net/model/sarvam-105b.md) | 49.5% | 35.4% | estimated ± 7.3 pp, low confidence |
| [LongCat-Flash-Lite-Sparse](https://benchgap.net/model/longcat-flash-lite-sparse.md) | 48.6% | 34.5% | estimated ± 7.3 pp, low confidence |
| [Sarvam 30B](https://benchgap.net/model/sarvam-30b.md) | 35.5% | 23.2% | estimated ± 7.3 pp, low confidence |

## Fitted on

| Model | BrowseComp | CyberGym |
|---|---|---|
| [Atria Dawn Preview](https://benchgap.net/model/atria-dawn-preview.md) | 92.5% | 86.5% |
| [Claude Mythos 5](https://benchgap.net/model/claude-mythos-5.md) | 88.0% | 83.8% |
| [Claude Opus 4.6](https://benchgap.net/model/claude-opus-4-6.md) | 83.7% | 66.6% |
| [Claude Opus 4.7](https://benchgap.net/model/claude-opus-4-7.md) | 79.3% | 73.1% |
| [DeepSeek V4 Flash 0731](https://benchgap.net/model/deepseek-v4-flash-0731.md) | 73.2% | 76.7% |
| [DeepSeek V4 Pro](https://benchgap.net/model/deepseek-v4-pro.md) | 83.4% | 83.3% |
| [GLM-5](https://benchgap.net/model/glm-5.md) | 62.0% | 43.2% |
| [GLM-5.1](https://benchgap.net/model/glm-5-1.md) | 68.0% | 68.7% |
| [GPT-5.4](https://benchgap.net/model/gpt-5-4.md) | 82.7% | 79.0% |
| [GPT-5.5](https://benchgap.net/model/gpt-5-5.md) | 84.4% | 81.8% |
| [GPT-5.6 Luna](https://benchgap.net/model/gpt-5-6-luna.md) | 83.3% | 77.9% |
| [GPT-5.6 Sol](https://benchgap.net/model/gpt-5-6-sol.md) | 92.2% | 84.5% |
| [GPT-5.6 Terra](https://benchgap.net/model/gpt-5-6-terra.md) | 87.5% | 81.8% |
| [Ling 3.1 Flash](https://benchgap.net/model/ling-3-1-flash.md) | 91.7% | 87.9% |
| [Step 5 Preview](https://benchgap.net/model/step-5-preview.md) | 88.7% | 84.7% |
