# Terminal-Bench 4.1 → Terminal-Bench 2.1 (Vals AI)

Calibration · https://benchgap.net/calibration/terminal-bench41/current/to/vals-terminal-bench21/current · data retrieved 2026-11-18

Terminal-Bench 2.1 (Vals AI) is estimated from Terminal-Bench 4.1 with a Michaelis–Menten + offset curve fitted on 32 models measured on both: y = 0.5020 + 0.3353·x / (0.07787 + x), R² = 0.52, cross-validated error 8.3 pp. It is used for 2 estimates.

| Estimated model | Terminal-Bench 4.1 | Terminal-Bench 2.1 (Vals AI) | Source |
|---|---|---|---|
| [GPT-6.1 Sol](https://benchgap.net/model/gpt-6-1-sol.md) | 60.9% | 79.9% | estimated ± 8.3 pp, medium confidence |
| [Hy4](https://benchgap.net/model/hy4.md) | 37.9% | 78.0% | estimated ± 8.3 pp, medium confidence |

## Fitted on

| Model | Terminal-Bench 4.1 | Terminal-Bench 2.1 (Vals AI) |
|---|---|---|
| [Claude Fable 5](https://benchgap.net/model/claude-fable.md) | 44.9% | 80.5% |
| [Claude Fable 5.1](https://benchgap.net/model/claude-fable-5-1.md) | 57.6% | 85.0% |
| [Claude Opus 4.8](https://benchgap.net/model/claude-opus-4-8.md) | 23.6% | 71.9% |
| [Claude Opus 5](https://benchgap.net/model/claude-opus-5.md) | 53.3% | 84.6% |
| [Claude Opus 5.5](https://benchgap.net/model/claude-opus-5-5.md) | 65.8% | 87.6% |
| [Claude Sonnet 5](https://benchgap.net/model/claude-sonnet-5.md) | 12.1% | 74.5% |
| [Claude Sonnet 5.5](https://benchgap.net/model/claude-sonnet-5-5.md) | 63.9% | 83.1% |
| [DeepSeek V4.1 Flash 1017](https://benchgap.net/model/deepseek-v4-1-flash.md) | 38.5% | 74.5% |
| [DeepSeek V4 Pro](https://benchgap.net/model/deepseek-v4-pro.md) | 6.7% | 54.7% |
| [Gemini 3.1 Pro](https://benchgap.net/model/gemini-3-1-pro.md) | 2.7% | 70.8% |
| [Gemini 3.7 Flash](https://benchgap.net/model/gemini-3-7-flash.md) | 12.4% | 77.5% |
| [Gemini 3.8 Flash](https://benchgap.net/model/gemini-3-8-flash.md) | 18.2% | 81.3% |
| [Gemini 4 Argon](https://benchgap.net/model/gemini-4-argon.md) | 55.5% | 57.7% |
| [GLM-5.3](https://benchgap.net/model/glm-5-3.md) | 41.2% | 71.5% |
| [GLM-5.3-Flash](https://benchgap.net/model/glm-5-3-flash.md) | 37.0% | 62.9% |
| [GPT-5.4 mini](https://benchgap.net/model/gpt-5-4-mini.md) | 1.8% | 54.7% |
| [GPT-5.6 Luna](https://benchgap.net/model/gpt-5-6-luna.md) | 17.6% | 79.0% |
| [GPT-5.6 Sol](https://benchgap.net/model/gpt-5-6-sol.md) | 38.5% | 85.8% |
| [GPT-5.6 Terra](https://benchgap.net/model/gpt-5-6-terra.md) | 21.5% | 77.5% |
| [GPT-6 Astra](https://benchgap.net/model/gpt-6-astra.md) | 60.3% | 87.3% |
| [GPT-6 Luna](https://benchgap.net/model/gpt-6-luna.md) | 15.2% | 73.0% |
| [GPT-6 Sol](https://benchgap.net/model/gpt-6-sol.md) | 50.3% | 83.1% |
| [Grok 4.5](https://benchgap.net/model/grok-4-5.md) | 12.4% | 67.8% |
| [Grok 4.6](https://benchgap.net/model/grok-4-6.md) | 19.7% | 78.3% |
| [Grok 4.7](https://benchgap.net/model/grok-4-7.md) | 38.8% | 73.4% |
| [Inkling](https://benchgap.net/model/inkling.md) | 0.0% | 47.6% |
| [Inkling-Small](https://benchgap.net/model/inkling-small.md) | 0.9% | 55.1% |
| [Kimi K3](https://benchgap.net/model/kimi-k3.md) | 23.0% | 80.9% |
| [MiMo-V2.6-Flash](https://benchgap.net/model/mimo-v2-6-flash.md) | 16.1% | 76.4% |
| [MiMo-V2.6-Pro](https://benchgap.net/model/mimo-v2-6-pro.md) | 27.0% | 67.8% |
| [MiniMax M3](https://benchgap.net/model/minimax-m3.md) | 7.6% | 53.6% |
| [Muse Spark 1.3](https://benchgap.net/model/muse-spark-1-3.md) | 14.9% | 72.3% |
