Skip to content
AIMODELRACE
Math · sub-raceNarrow lead

Math Chat race

Solving and explaining math in conversation, judged by blind votes on math prompts.

Leader
Gemini 4 Argon
Margin
8 Elo
0.5% ahead
Weeks at #1
4
in a row
Changes
10
in 2026

Warming up the track…

Oct 20, 2026Gemini 4 Argon leads
2025Apr ’25Jul ’25Oct ’252026Apr ’26Jul ’26Oct ’26

Score history

Top 6 today over the leader's frontier · click to jump to a week

#1 frontierGemini 4 ArgonFable 5.13.8 FlashOpus 5.5GLM-5.3 FlashQwen3.8-Max
1,2501,3001,3501,4001,4501,5001,5501,6002025Apr ’25Jul ’25Oct ’252026Apr ’26Jul ’26Oct ’26

Reigns at #1

Longest: Opus 5, 10 weeks (from Jul 2026).

Sibling races

More math races

Why this race matters

Most people meet math through homework, finance or work questions in chat, not competitions.

How it is scored

LMArena Text leaderboard, Math category: ratings from blind pairwise votes.

Unit: eloHigher is betterPhoto finish under 5 EloDominant over 30 Elo

Data provenance

  • LMArena Text · Math ↗

    Blind pairwise votes, reported as Bradley-Terry ratings on an Elo scale. Data: lmarena-ai/leaderboard-dataset, CC BY 4.0.

LMArena Text · Math · LMArena

Weekly history from the source's published leaderboards.

68 tracked models with a result (400 entries on the source board) · newest data October 15, 2026.

Sources, licences and credits