# Vals SWE-bench → SWE-bench Pro

Calibration · https://benchgap.net/calibration/1599 · data retrieved 2026-10-29

SWE-bench Pro is estimated from Vals SWE-bench with a Michaelis–Menten + offset curve fitted on 35 models measured on both: y = 0.0982 + 2.0000·x / (2.41398 + x), R² = 0.49, cross-validated error 5.4 pp. It is used for 0 estimates.

## Fitted on

| Model | Vals SWE-bench | SWE-bench Pro |
|---|---|---|
| [Claude Fable 5](https://benchgap.net/model/claude-fable.md) | 95.0% | 80.0% |
| [Claude Opus 4.8](https://benchgap.net/model/claude-opus-4-8.md) | 88.6% | 69.2% |
| [Claude Opus 5](https://benchgap.net/model/claude-opus-5.md) | 97.0% | 79.2% |
| [Claude Sonnet 5](https://benchgap.net/model/claude-sonnet-5.md) | 79.6% | 63.2% |
| [DeepSeek V4 Flash 0731](https://benchgap.net/model/deepseek-v4-flash-0731.md) | 88.8% | 52.6% |
| [DeepSeek V4 Pro 0813](https://benchgap.net/model/deepseek-v4-pro-0813.md) | 96.4% | 55.4% |
| [Gemini 3.5 Flash](https://benchgap.net/model/gemini-3-5-flash.md) | 78.8% | 55.1% |
| [Gemini 3.5 Flash-Lite](https://benchgap.net/model/gemini-3-5-flash-lite.md) | 75.0% | 54.2% |
| [GLM-5.1](https://benchgap.net/model/glm-5-1.md) | 76.4% | 58.4% |
| [GLM-5.2](https://benchgap.net/model/glm-5-2.md) | 82.8% | 62.1% |
| [GPT-5.3 Codex](https://benchgap.net/model/gpt-5-3-codex.md) | 78.0% | 56.8% |
| [GPT-5.5](https://benchgap.net/model/gpt-5-5.md) | 82.6% | 58.6% |
| [GPT-5.6 Luna](https://benchgap.net/model/gpt-5-6-luna.md) | 93.0% | 62.7% |
| [GPT-5.6 Sol](https://benchgap.net/model/gpt-5-6-sol.md) | 96.2% | 64.6% |
| [GPT-5.6 Terra](https://benchgap.net/model/gpt-5-6-terra.md) | 95.4% | 63.4% |
| [Grok 4.20](https://benchgap.net/model/grok-4-20-beta.md) | 72.2% | 51.8% |
| [Grok 4.5](https://benchgap.net/model/grok-4-5.md) | 86.6% | 64.7% |
| [Inkling](https://benchgap.net/model/inkling.md) | 77.6% | 54.3% |
| [Inkling-Small](https://benchgap.net/model/inkling-small.md) | 82.2% | 55.9% |
| [Kimi K2.6](https://benchgap.net/model/kimi-2-6.md) | 76.2% | 58.6% |
| [Laguna M.1](https://benchgap.net/model/laguna-m-1.md) | 57.6% | 49.2% |
| [Laguna XS.2](https://benchgap.net/model/laguna-xs-2.md) | 55.2% | 46.3% |
| [Ling 3.0 Flash](https://benchgap.net/model/ling-3-0-flash.md) | 65.2% | 56.6% |
| [MiMo-V2.5](https://benchgap.net/model/mimo-v2-5.md) | 71.0% | 56.1% |
| [MiMo-V2.5-Pro](https://benchgap.net/model/mimo-v2-5-pro.md) | 74.0% | 57.2% |
| [MiniMax M2.7](https://benchgap.net/model/minimax-m2-7.md) | 73.8% | 56.2% |
| [MiniMax M3](https://benchgap.net/model/minimax-m3.md) | 75.0% | 59.0% |
| [Muse Spark](https://benchgap.net/model/muse-spark.md) | 74.4% | 52.4% |
| [Muse Spark 1.1](https://benchgap.net/model/muse-spark-1-1.md) | 82.0% | 61.5% |
| [Qwen3.6-27B](https://benchgap.net/model/qwen3-6-27b.md) | 70.0% | 53.5% |
| [Qwen 3.6 Max (preview)](https://benchgap.net/model/qwen3-6-max-preview.md) | 72.8% | 57.3% |
| [Qwen3.6 Plus](https://benchgap.net/model/qwen3-6-plus.md) | 73.4% | 56.6% |
| [Qwen3.7 Max](https://benchgap.net/model/qwen3-7-max.md) | 68.8% | 60.6% |
| [Qwen3.8-27B](https://benchgap.net/model/qwen3-8-27b.md) | 86.0% | 61.7% |
| [Qwen3.8 Max](https://benchgap.net/model/qwen3-8-max.md) | 85.6% | 67.7% |