Vals Index leaderboard (Oct 2026): AI model scores
As of 21 Oct 2026, the highest reported xAI score on Vals Index is Grok 4.7 (xAI) at 58.56%, #11 on that leaderboard, per Vals AI.
What it measures
Agentic performance on finance, coding, legal and tax tasks, weighted by each sector’s share of U.S. GDP.
One score for agentic work across finance, coding, legal and tax tasks, weighted by each sector’s share of US GDP.
What the score means
Accuracy: weighted average of 8 component benchmarks
Vals’ weighted average of model performance across its finance, coding, legal and tax benchmarks, each sector weighted by its share of U.S. GDP.
Results from published AI leaderboards. Compare models within each benchmark; scores from different tests aren’t directly comparable. How we collect these results
Scores by model
Run by Vals AI. Leaderboard dated 21 Oct 2026. Last checked 22 Oct 2026.
Highlight a provider: Anthropic OpenAI Google xAI Meta DeepSeek Z.ai Moonshot MiniMax Xiaomi
Highlighting xAI models. Other providers are dimmed. Show all providers
- Anthropic
- OpenAI
- xAI
- Meta
- DeepSeek
- Z.ai
- Moonshot
- MiniMax
- Xiaomi
- Other
- Accuracy
- Vals’ weighted average of model performance across its finance, coding, legal and tax benchmarks, each sector weighted by its share of U.S. GDP.
- Cost per test (USD)
- Average API cost to run the index tasks, as published.
All scores
| Rank | Model | Provider | Accuracy | Cost per test (USD) | Price in / out ($/1M tokens) |
|---|---|---|---|---|---|
| 1 | Claude Sonnet 5.5 | Anthropic | 70.85% | $19 | $2 / $10 |
| 2 | Claude Opus 5.5 | Anthropic | 70.18% | $28 | $4 / $20 |
| 3 | Gemini 4 Argon | 69.24% | $13 | $4 / $20 | |
| 4 | Wonder Full | Meta | 66.79% | $6.81 | $4 / $20 |
| 5 | GPT-6 Astra | OpenAI | 65.02% | $21 | $10 / $50 |
| 6 | Wonder | Meta | 64.88% | $11 | $4 / $20 |
| 7 | GPT-6.1 Sol | OpenAI | 63.94% | $3.61 | $2 / $10 |
| 8 | DeepSeek V4.1 Flash 1017 | DeepSeek | 61.26% | $0.66 | $0.3 / $1.2 |
| 9 | GLM 5.3 | Z.ai | 59.73% | $8.52 | $1.4 / $4.4 |
| 10 | Claude Haiku 5.5 | Anthropic | 58.71% | $3.17 | $0.1 / $0.5 |
Show 14 more rows
| Rank | Model | Provider | Accuracy | Cost per test (USD) | Price in / out ($/1M tokens) |
|---|---|---|---|---|---|
| 11 | Grok 4.7 | xAI | 58.56% | $12 | $2 / $6 |
| 12 | MiMo V2.6 Pro | Xiaomi | 57.81% | $0.43 | $0.435 / $0.87 |
| 13 | Gemini 3.8 Flash | 55.95% | $6.06 | $1.5 / $7.5 | |
| 14 | MiMo V2.6 Flash | Xiaomi | 54.83% | $0.22 | $0.14 / $0.28 |
| 15 | MiniMax-M3.1-Flash | MiniMax | 53.78% | $0.66 | $0.15 / $0.6 |
| 16 | Hy4 Preview | Other | 53.21% | $1.45 | $0.834 / $2.501 |
| 17 | DeepSeek V4 Pro 0813 | DeepSeek | 52.97% | $2.50 | $1.32 / $3.96 |
| 18 | Kimi K3 | Moonshot | 52.95% | $7.17 | $3 / $15 |
| 19 | GPT-6 Luna | OpenAI | 52.32% | $0.45 | $0.1 / $0.5 |
| 20 | GLM 5.3 Flash | Z.ai | 50.86% | $1.13 | $0.075 / $0.25 |
| 21 | Muse Spark Open | Meta | 48.57% | $4.72 | $1.25 / $4.25 |
| 22 | Beam 1.0 | Other | 43.42% | – | N/A |
| 23 | Inkling | Other | 28.96% | $1.85 | $1 / $4.05 |
| 24 | Inkling Small | Other | 27.59% | $0.40 | $0.3 / $1.2 |
Source: Vals AI leaderboard ↗
Best-ranked xAI model here: Grok 4.7 (xAI). Try prompts for Grok Bot →