LMARENA ARENA SCORE TEXT STYLE-CONTROL · OVERALL

Anthropic leads. Meta ranks second. Published July 12, 2026.

In LMArena’s official text style-control overall snapshot, claude-fable-5 leads at 1507.5 and muse-spark-1.1 ranks second at 1491.6. We select the highest-scoring model for each official organization. This is a point-in-time table with confidence intervals and battle counts. It is not spliced into the separate May 2023 to August 2025 legacy archive.

See current standings Audit the method

Chatbot Arena archive

Sampled provider frontier ratings, May 2023 to August 2025

Meta highlighted. 827 to 1288 published rating; sampled peak 1396.

Eleven selected published snapshots are shown. LMArena changed its rating method during this window, including the move from Elo to Bradley-Terry in January 2024 and later style-control and frequency-reweighting changes. The selected snapshots cross LMArena rating-method changes, including Bradley-Terry, style control, and frequency reweighting. Endpoint differences are not normalized like-for-like performance changes. This legacy series is never joined to the current table. Each raw snapshot has a published SHA-256 receipt.

Current LMArena snapshot

Four facts define the current board.

These values come from LMArena’s text style-control overall table published 2026-07-12. The legacy archive remains separate.

1507.5

Anthropic leads

Claude Fable 5 is the highest-scoring model in the official snapshot, with a reported interval of 1500.0 to 1515.0.

15.9

Gap to second

Meta’s Muse Spark 1.1 ranks second at 1491.6, 15.9 points behind Anthropic.

7,856

Leader battles

LMArena reports 7,856 battles for Claude Fable 5 in this published snapshot.

Tracked organizations

Every row is selected by official organization metadata, not a model-name substring.

Point-in-time standings

Current provider frontier, with uncertainty shown.

Provider frontier means the highest Arena Score for each official organization in this one published snapshot. It is not a composite score and does not continue the historical chart.

LMArena Arena Score, text style-control, overall category, published 2026-07-12. Higher is better.
Provider rank	Model	Organization	Arena Score	Reported interval	Battles
01	claude-fable-5	Anthropic	1507.5	1500.0– 1515.0	7,856
02	muse-spark-1.1	Meta	1491.6	1482.5– 1500.7	4,498
03	gemini-3-pro	Google	1485.9	1482.0– 1489.7	41,290
04	gpt-5.6-sol-xhigh	OpenAI	1484.6	1473.2– 1496.1	2,821
05	grok-4.20-beta1	xAI	1474.4	1469.7– 1479.0	26,877
06	deepseek-v4-pro	DeepSeek	1457.1	1452.6– 1461.6	41,800
07	mistral-medium-3.5	Mistral	1427.2	1420.6– 1433.8	11,034

Source revision: 4a6c22ada142f24563a2ecc5ceb1ce94ab8be019. Artifact SHA-256: 8884336d5d6307191bacac5f7bab560c9f597908f549b1b542b861f323319abe. The current table and legacy archive are separate and never blended.

First lab report

Meta climbed 461 points in the corrected legacy archive.

Open Meta’s race file

827 → 1396 → 1288

Meta’s pipeline-checked archive path

The corrected provider mapper excludes NVIDIA Nemotron, Tulu, SteerLM, and other third-party Llama derivatives.

Chart lab

Gap chase, reign tape, head to head

Inspect three chart modes →

Methodology

What is measured, and what is not

Read the scoring contract →

Direct answers

How Model Gauntlet works.

What is Model Gauntlet?

Model Gauntlet publishes sourced AI model standings and a separately labeled historical race archive. Current and legacy metrics are never blended.

How current are the standings?

The current provider table uses LMArena's official text style-control overall snapshot published July 12, 2026. It includes confidence intervals and battle counts.

Does the current table continue the historical chart?

No. The current style-control Arena Score and the legacy May 2023 to August 2025 archive use incompatible methodology and inclusion rules. No cross-window change is calculated.

Race alerts preview

Join the future alert list.

Alert delivery is not active yet. The form is a preview until submission receipt and outbound delivery are verified.