BenchLeader
Menu
Best LLMsChoose a modelCoding plansLeaderboardsModelsBenchmarksProvidersReleases7

LMArena Maths leaderboard

As of 6 Nov 2026, the top score on LMArena Maths is 1533 by Gemini 4 Argon (Google), ahead of Claude Fable 5.1 at 1528, across 365 model configurations with a published result (Arena (LMArena), result dated 28 Oct 2026). Pairwise votes on prompts classified as mathematics. LMArena Maths counts toward the BenchLeader maths score.

Key facts (as of 6 Nov 2026)

  • Gemini 4 Argon holds the top Arena Maths score at 1533 (Arena (LMArena), 28 Oct 2026).
  • Claude Fable 5.1 is second on Arena Maths at 1528 (Arena (LMArena), 28 Oct 2026).
  • Arena Maths: Text arena rating on maths prompts (Arena (LMArena)).
  • 365 model configurations have a published Arena Maths result on BenchLeader (BenchLeader, 6 Nov 2026).
  • Arena Maths feeds the maths category of the BenchLeader Index (BenchLeader methodology).

Text arena rating on maths prompts.

Published by
Arena (LMArena)
Category
Maths
Index weight
1.0
Models
365
Data as of
6 Nov 2026

CC BY 4.0 — lmarena-ai/leaderboard-dataset on Hugging Face.

What the test looks like

Pairwise votes on prompts classified as mathematics.

How it is scored

Bradley-Terry rating from pairwise votes, in Elo-like units, with style control.

What to keep in mind

Voters judge the answer as written; correctness is not verified. Counts toward the maths category since index v1.2.

Top 15
  1. 1Gemini 4 Argon (high)1533
  2. 2Claude Fable 5.1 (max)1528
  3. 3Gemini 4 Carbon (high)1525
  4. 4Gemini 3.8 Flash (high)1524
  5. 5Claude Opus 5 (high)1521
  6. 6Claude Fable 5 (high)1521
  7. 7Claude Opus 4.6 (high)1517
  8. 8Claude Opus 5 (max)1515
  9. 9Claude Opus 5.5 (high)1512
  10. 10GPT-6.1 Sol (max)1509
  11. 11Muse Spark 1.3 (max)1508
  12. 12Claude Sonnet 5.5 (xhigh)1508
  13. 13Claude Opus 4.7 (high)1504
  14. 14Gemini 3.6 Flash (high)1504
  15. 15Claude Opus 4.61503
365 of 365
#
1Gemini 4 Argonhigh153366.02026-10-28
2Claude Fable 5.1max152867.22026-10-28
3Gemini 4 Carbonhigh152569.22026-10-28
4Gemini 3.8 Flashhigh152462.02026-10-28
5Claude Opus 5high152164.32026-10-28
6Claude Fable 5high152164.32026-10-28
7Claude Opus 4.6high151762.62026-10-28
8Claude Opus 5max151565.42026-10-28
9Claude Opus 5.5high151265.82026-10-28
10GPT-6.1 Solmax150967.32026-10-28
11Muse Spark 1.3max150862.02026-10-28
12Claude Sonnet 5.5xhigh150863.42026-10-28
13Claude Opus 4.7high150460.92026-10-28
14Gemini 3.6 Flashhigh150461.22026-10-28
15Claude Opus 4.6150362.32026-10-28
16Qwen3.8 Maxmax150358.82026-10-28
17GLM 5.3 Flash150356.82026-10-28
18Gemini 3.7 Flashhigh150362.22026-10-28
19GPT-5.5150061.72026-10-28
20GLM 5.3max149959.02026-10-28
21Kimi K3max149962.12026-10-28
22DeepSeek V4.1 Flashmax149859.02026-10-28
23Gemini 3.5 Flashhigh149762.52026-10-28
24GPT-5.6 Solxhigh149564.22026-10-28
25Muse Spark 1.1149462.22026-10-28
26GPT-5.5high149363.72026-10-28
27Claude Opus 4.8high149360.62026-10-28
28Muse Spark 1.2xhigh149261.72026-10-28
29Claude Opus 4.7149163.12026-10-28
30GPT-5.4high149158.82026-10-28
31MiMo-V2.6-Pro149057.32026-10-28
32Muse Wondermax148962.32026-10-28
33GPT-6 Astramax148967.82026-10-28
34Gemini 3.1 Pro148860.52026-10-28
35GLM 5.2max148759.52026-10-28
36Hy3148555.82026-10-28
37Qwen3.7max148260.32026-10-28
38Gemini 3.5 Flashmedium148161.92026-10-28
39GPT-5.6 Terraxhigh148060.32026-10-28
40Inkling148047.52026-10-28
41MiMo-V2.5-Pro147954.92026-10-28
42Kimi K2.6147955.22026-10-28
43ERNIE 5.11478–2026-10-28
44GPT-5.6 Lunaxhigh147758.52026-10-28
45Gemini 3 Pro147760.22026-10-28
46Claude Opus 4.8147760.02026-10-28
47Beam 1.0 501B A23B147754.52026-10-28
48Qwen3.6 Maxmax147662.42026-10-28
49GLM 5.1147555.72026-10-28
50Gemini 3 Flash147556.92026-10-28
51Claude Sonnet 5high147559.32026-10-28
52Grok 4.5147259.82026-10-28
53Claude Opus 4.5high147259.02026-10-28
54Kimi K2.5thinking147158.72026-10-28
55Gemma 4 31B146947.82026-10-28
56Gemma 4 26B A4B146956.12026-10-28
57Qwen3.5 Maxmax1468–2026-10-28
58Mistral Large 4146852.92026-10-28
59Qwen3.7 Plus146757.72026-10-28
60Muse Spark146664.12026-10-28

More maths benchmarks · All LLM benchmarks

Related benchmarks

Top head-to-heads on Arena Maths

Cite as: BenchLeader, “Arena Maths leaderboard”, data as of 6 Nov 2026, https://www.benchleader.com/benchmarks/lmarena_text_math.

Arena Maths: questions

What does Arena Maths measure?
Pairwise votes on prompts classified as mathematics. Scores are reported in Elo-style ratings from pairwise comparisons; higher is better.
What is the top score on LMArena Maths?
As of 6 Nov 2026, the highest score on Arena Maths is 1533 by Gemini 4 Argon (result dated 28 Oct 2026), ahead of Claude Fable 5.1 at 1528. Scores come from Arena (LMArena) and are re-read daily.
Which AI model leads Arena Maths?
Gemini 4 Argon leads Arena Maths with 1533 as of 6 Nov 2026 (result dated 28 Oct 2026), ahead of Claude Fable 5.1 at 1528.
How many models have Arena Maths results?
365 model configurations have a Arena Maths result on BenchLeader, all taken from Arena (LMArena).
Who runs Arena Maths and how often is it updated?
Arena Maths is published by Arena (LMArena). BenchLeader re-reads the published results once a day and records the date each result was published.
Does Arena Maths count toward the BenchLeader Index?
Yes. Arena Maths contributes to the maths category of the BenchLeader Index, normalised so that 50 is the average of the evaluated models.