Vals Router Index
Long agentic work in finance, coding, law and tax, where model routers are compared with single models. Run by Vals AI.
As of 24 Oct 2026, Gemini 4 Argon leads Vals Router Index on BenchLeader with 69.2%, ahead of Claude Sonnet 5.5 at 69.0%, across 20 model configurations with a published result.
- Published by
- Vals AI
- Category
- Composite
- Index weight
- Reference only
- Models
- 20
- Data as of
- 24 Oct 2026
- Newest result
- 20 Oct 2026
Vals AI (vals.ai).
- 1Gemini 4 Argon69.2%
- 2Claude Sonnet 5.569.0%
- 3Muse Wonder Full66.8%
- 4GPT-6 Astra65.0%
- 5Muse Wonder64.9%
- 6Claude Opus 5.564.2%
- 7GPT-6.1 Sol63.9%
- 8DeepSeek V4.1 Flash 101761.3%
- 9GLM 5.359.7%
- 10Claude Haiku 5.558.7%
- 11Grok 4.758.6%
- 12MiMo-V2.6-Pro57.8%
- 13Gemini 3.8 Flash56.0%
- 14MiMo-V2.6-Flash54.8%
- 15MiniMax M3.1 Flash53.8%
20 of 20
| # | ||||
|---|---|---|---|---|
| 1 |  Gemini 4 Argon | 69.2% | 64.9 | 2026-10-20 |
| 2 |  Claude Sonnet 5.5 | 69.0% | – | 2026-10-20 |
| 3 |  Muse Wonder Full | 66.8% | – | 2026-10-20 |
| 4 |  GPT-6 Astra | 65.0% | 68.0 | 2026-10-20 |
| 5 | Muse Wonder | 64.9% | 63.7 | 2026-10-20 |
| 6 | Claude Opus 5.5 | 64.2% | 62.0 | 2026-10-20 |
| 7 | GPT-6.1 Sol | 63.9% | 63.0 | 2026-10-20 |
| 8 | New DeepSeek V4.1 Flash 1017open ↗ | 61.3% | – | 2026-10-20 |
| 9 |  GLM 5.3open ↗ | 59.7% | – | 2026-10-20 |
| 10 | Claude Haiku 5.5 | 58.7% | – | 2026-10-20 |
| 11 |  Grok 4.7 | 58.6% | – | 2026-10-20 |
| 12 |  MiMo-V2.6-Proopen ↗ | 57.8% | 64.0 | 2026-10-20 |
| 13 | Gemini 3.8 Flash | 56.0% | 55.7 | 2026-10-20 |
| 14 | MiMo-V2.6-Flashopen ↗ | 54.8% | 59.7 | 2026-10-20 |
| 15 | New MiniMax M3.1 Flashopen ↗ | 53.8% | 62.0 | 2026-10-20 |
| 16 | DeepSeek V4 Proopen ↗ | 53.0% | 53.5 | 2026-10-20 |
| 17 |  Kimi K3open ↗ | 53.0% | 59.1 | 2026-10-20 |
| 18 | GPT-6 Luna | 52.3% | – | 2026-10-20 |
| 19 | GLM 5.3 Flashopen ↗ | 50.9% | 62.6 | 2026-10-20 |
| 20 | Inkling Smallopen ↗ | 27.6% | 54.8 | 2026-10-20 |
More composite benchmarks · All LLM benchmarks
Cite as: BenchLeader, “Vals Router Index leaderboard”, data as of 24 Oct 2026, https://www.benchleader.com/benchmarks/vals_router_index.
Vals Router Index: questions
- What does Vals Router Index measure?
- Long agentic work in finance, coding, law and tax, where model routers are compared with single models. Run by Vals AI. Scores are reported in percent of tasks solved; higher is better.
- Which AI model leads Vals Router Index?
- Gemini 4 Argon leads Vals Router Index with 69.2% as of 24 Oct 2026, ahead of Claude Sonnet 5.5 at 69.0%.
- How many models have Vals Router Index results?
- 20 model configurations have a Vals Router Index result on BenchLeader, all taken from Vals AI.
- Who runs Vals Router Index and how often is it updated?
- Vals Router Index is published by Vals AI. BenchLeader re-reads the published results once a day and records the date each result was published.
- Does Vals Router Index count toward the BenchLeader Index?
- No. Vals Router Index is shown for reference but left out of the composite index.