Vals Multimodal Index
Indexes · 2025-11-19
Composite index weighting finance, coding, and education tasks.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Fable 5 | 74.2 |
| Claude Opus 5 | 73.9 |
| Kimi K3 | 73.4 |
| GPT-5.6 Sol | 72.2 |
| Opus 4.8 | 70.9 |
| GPT-5.6 Luna | 69.1 |
| Sonnet 5 | 68.8 |
| GPT-5.5 | 68.1 |
| Opus 4.7 | 67.4 |
| Muse Spark 1.1 | 66.7 |
| GPT-5.6 Terra | 65.1 |
| Grok 4.5 | 63.4 |
| Gemini 3.5 Flash | 62.9 |
| Gemini 3.6 Flash | 62.8 |
| Sonnet 4.6 | 60.6 |
| MiniMax M3 | 60.0 |
| Kimi K2.6 | 56.4 |
| Gemini 3.1 Pro Preview | 56.1 |
| GPT-5.4 Mini | 54.2 |
| Qwen3.7-Plus | 53.9 |
| MiMo-V2.5 | 52.8 |
| Gemini 3 Flash Preview | 52.2 |
| Qwen3.6 Plus (2026-04-02) | 51.5 |
| GPT-5.4 Nano | 47.6 |
| Grok 4.3 | 43.3 |
| Haiku 4.5 | 42.9 |
| Gemini 3.1 Flash-Lite Preview | 41.4 |
| Grok 4.20 | 39.1 |
| Mistral Medium 3.5 | 34.8 |