Vals Index leaderboard
- Claude Sonnet 5.5 leads
- 70.8%
- models measured
- 21
- lead over #2
- +0.7 pts
- median
- 57.8%
Leader on Vals Index · read October 22
On October 22, 2026, 21 of our catalog's models had a Vals Index result, and Anthropic Claude Sonnet 5.5 led with 70.8%, ahead of Anthropic Claude Opus 5.5 at 70.2%.
The Vals Index from vals.ai is a composite over many kinds of professional tasks. It measures general capability, not coding.
Role in the index
Vals Index is shown as published and does not feed any of our indexes. Read the methodology
| # | Model | Result | Cost per test | Duration, as printed | Evidence | Coding index | Output $/Mtok |
|---|---|---|---|---|---|---|---|
| 1 | Claude Sonnet 5.5Anthropic | 70.8% | $18.70 | 1h 26m | 87.3 | $10.00 | |
| 2 | Claude Opus 5.5Anthropic | 70.2% | $28.12 | 1h 25m | 88.8 | $20.00 | |
| 3 | Gemini 4 ArgonGoogle | 69.2% | $13.37 | 44m 58s | 81.0 | $10.00 | |
| 4 | GPT-6 AstraOpenAI | 65.0% | $21.28 | 33m 39s | 85.1 | $50.00 | |
| 5 | Muse WonderMeta | 64.9% | $10.65 | 35m 24s | 71.9 | $20.00 | |
| 6 | GPT-6.1 SolOpenAI | 63.9% | $3.61 | 44m 24s | 82.4 | $10.00 | |
| 7 | DeepSeek V4.1 Flash 1017DeepSeek | 61.3% | $0.66 | 39m 55s | 70.3 | $0.60 | |
| 8 | GLM 5.3Z.ai | 59.7% | $8.52 | 1h 04m | 70.6 | $4.40 | |
| 9 | Claude Haiku 5.5Anthropic | 58.7% | $3.17 | 49m 48s | 70.7 | $0.50 | |
| 10 | Grok 4.7xAI | 58.6% | $11.63 | 33m 34s | 66.7 | $6.00 |
Only models in our catalog are listed, and each appears once with its best published run. 21 rows were read from the source on October 22, 2026, and rows that match no catalog model are left out. Cost per test is what Vals AI reports one test cost on the run behind the result, in US dollars. Every other column is a figure the board publishes for the same run.
Cheapest good model on Vals Index · Best open-weight model on Vals Index