Benchmarks / Composite
Vals Multimodal Index v1.2
Benchmark consisting of a weighted performance across finance, coding, and education tasks. Showing the potential impact that LLM's can have on the economy. Scores are the values published by VALS AI. Source benchmark version 1.2.
Vals Multimodal Index v1.2
Higher is better · best effort per model
full table below
| # | Model | Effort | Score |
|---|---|---|---|
| 1 | Fable 5 | max | 74.2% |
| 2 | Opus 5 | max | 73.9% |
| 3 | Kimi K3 | max | 73.4% |
| 4 | GPT-5.6 Sol | max | 72.6% |
| 5 | Opus 4.8 | max | 70.9% |
| 6 | Muse Spark 1.2 | max | 69.8% |
| 7 | GPT-5.6 Luna | max | 69.1% |
| 8 | Sonnet 5 | max | 68.8% |
| 9 | GPT-5.5 | xhigh | 68.1% |
| 10 | Opus 4.7 | high | 67.4% |
| 11 | Muse Spark 1.1 | xhigh | 66.7% |
| 12 | Qwen 3.8 Max | max | 65.4% |
| 13 | Gemini 3.6 Flash | high | 65.1% |
| 14 | GPT-5.6 Terra | xhigh | 65.1% |
| 15 | Grok 4.5 | high | 63.4% |
| 16 | Gemini 3.5 Flash | high | 62.9% |
| 17 | Sonnet 4.6 | max | 60.6% |
| 18 | MiniMax M3 | max | 60.0% |
| 19 | Kimi K2.6 | max | 56.4% |
| 20 | Gemini 3.1 Pro | high | 56.1% |
| 21 | GPT-5.4 Mini | xhigh | 54.2% |
| 22 | Qwen 3.7 Plus | high | 53.9% |
| 23 | MiMo V2.5 | xhigh | 52.8% |
| 24 | Gemini 3 Flash | high | 52.2% |
| 25 | Qwen 3.6 Plus | max | 51.5% |
| 26 | Inkling Small | xhigh | 50.1% |
| 27 | Inkling | max | 49.4% |
| 28 | GPT-5.4 Nano | high | 47.6% |
| 29 | Grok 4.3 | high | 43.3% |
| 30 | Haiku 4.5 | xhigh | 42.9% |
| 31 | Gemini 3.1 Flash-Lite | high | 41.4% |
| 32 | Grok 4.20 | xhigh | 39.1% |
| 33 | Mistral Medium 3.5 | high | 34.8% |