Roboflow Vision Evals — Reasoning, Judged (September)
Multimodal
Visual reasoning correctness judged against ground truth by Gemini 3.5 Flash at temperature zero. One or three runs per effort tier; atomic sample count unreported.
Top models (higher is better)
| Model | Score |
|---|---|
| GPT-6 Astra | 91.2 |
| Claude Opus 5.5 | 85.9 |
| Gemini 3.8 Flash | 84.5 |
| Gemini 3.5 Flash | 82.1 |
| Gemini 3.7 Flash | 81.9 |
| Gemini 3.6 Flash | 81.0 |
| GPT-6 Sol | 77.9 |
| Muse Spark 1.1 | 76.6 |
| Muse Spark 1.2 | 75.7 |
| Gemini 3.1 Pro Preview | 74.8 |
| Gemini 3 Flash Preview | 74.2 |
| Claude Opus 5 | 74.2 |
| Kimi K3 | 74.2 |
| Muse Spark 1.3 | 73.3 |
| Claude Fable 5.1 | 73.1 |
| GPT-5.5 | 72.2 |
| GPT-5.6 Sol | 71.7 |
| Qwen3.8 Flash | 69.5 |
| Gemini 3.5 Flash-Lite | 68.9 |
| Qwen3.7-Plus | 68.2 |
| Grok 4.7 | 66.9 |
| Claude Fable 5 | 66.2 |
| GPT-5.6 Luna | 65.6 |
| GPT-5.6 Terra | 65.3 |
| GPT-5.4 Mini | 64.0 |
| Grok 4.6 | 63.8 |
| Muse Glimmer 30B | 62.9 |
| Gemini 2.5 Pro | 62.3 |
| Qwen3.7 Flash | 61.6 |
| Qwen3.5-27B | 61.6 |
| GPT-6 Luna | 60.7 |
| Grok 4.5 | 59.8 |
| GLM-5.3 Flash | 59.6 |
| Kimi K2.6 | 58.9 |
| MiMo-V2.6-Flash | 58.5 |
| MiMo-V2.6-Pro | 55.9 |
| Opus 4.8 | 53.0 |
| GLM 5V Turbo | 49.7 |
| Sonnet 5 | 43.0 |
| Qwen3-VL-235B-A22B-Instruct | 33.8 |