BenchLeader
Menu
Best LLMsChoose a modelLeaderboardsModelsBenchmarksProvidersReleases12

Vals Index

Vals AI's composite across its benchmarks. Shown for reference.

As of 24 Oct 2026, Claude Sonnet 5.5 leads Vals Index on BenchLeader with 70.8, ahead of Claude Opus 5.5 at 70.2, across 24 model configurations with a published result.

Published by
Vals AI
Category
Composite
Index weight
Reference only
Models
24
Data as of
24 Oct 2026
Newest result
22 Oct 2026

Vals AI (vals.ai).

What the test looks like

Vals AI's composite across the benchmarks it runs, from legal and finance to maths and coding.

How it is scored

An average of component scores on Vals AI's scale. Shown for reference only.

What to keep in mind

Its membership and weights are Vals AI's choices and change over time.

Top 15
  1. 1Claude Sonnet 5.570.8
  2. 2Claude Opus 5.570.2
  3. 3Gemini 4 Argon (high)69.2
  4. 4Muse Wonder Full (max)66.8
  5. 5GPT-6 Astra (max)65.0
  6. 6Muse Wonder (max)64.9
  7. 7GPT-6.1 Sol (max)63.9
  8. 8DeepSeek V4.1 Flash 1017 (max)61.3
  9. 9GLM 5.3 (max)59.7
  10. 10Qwen3.8 Max Internal59.0
  11. 11Claude Haiku 5.558.7
  12. 12Grok 4.7 (xhigh)58.6
  13. 13MiMo-V2.6-Pro57.8
  14. 14Gemini 3.8 Flash (high)56.0
  15. 15MiMo-V2.6-Flash54.8
24 of 24
#
1Claude Sonnet 5.570.8–2026-10-22
2Claude Opus 5.570.262.02026-10-22
3Gemini 4 Argonhigh69.270.82026-10-22
4Muse Wonder Fullmax66.870.72026-10-22
5GPT-6 Astramax65.070.42026-10-22
6Muse Wondermax64.967.52026-10-22
7GPT-6.1 Solmax63.969.92026-10-22
8NewDeepSeek V4.1 Flash 1017max61.364.02026-10-22
9GLM 5.3max59.763.52026-10-22
10NewQwen3.8 Max Internal59.0–2026-10-22
11Claude Haiku 5.558.7–2026-10-22
12Grok 4.7xhigh58.661.62026-10-22
13MiMo-V2.6-Pro57.864.02026-10-22
14Gemini 3.8 Flashhigh56.064.42026-10-22
15MiMo-V2.6-Flash54.859.72026-10-22
16NewMiniMax M3.1 Flash53.862.02026-10-22
17DeepSeek V4 Promax53.060.82026-10-22
18Kimi K3max53.064.72026-10-22
19GPT-6 Lunamax52.360.02026-10-22
20GLM 5.3 Flashmax50.954.22026-10-22
21Muse Spark Openxhigh48.662.02026-10-22
22NewMercury 3high43.7–2026-10-22
23NewBeam 1.043.451.12026-10-22
24Inkling Small27.654.82026-10-22

More composite benchmarks · All LLM benchmarks

Cite as: BenchLeader, “Vals Index leaderboard”, data as of 24 Oct 2026, https://www.benchleader.com/benchmarks/vals_index.

Vals Index: questions

What does Vals Index measure?
Vals AI's composite across the benchmarks it runs, from legal and finance to maths and coding. Scores are reported in index; higher is better.
Which AI model leads Vals Index?
Claude Sonnet 5.5 leads Vals Index with 70.8 as of 24 Oct 2026, ahead of Claude Opus 5.5 at 70.2.
How many models have Vals Index results?
24 model configurations have a Vals Index result on BenchLeader, all taken from Vals AI.
Who runs Vals Index and how often is it updated?
Vals 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 Index count toward the BenchLeader Index?
No. Vals Index is shown for reference but left out of the composite index.