Anthropic · proprietary on Arena · first on Arena 6 February 2026

claude-opus-4-6

72 days in first place

It held first place on Arena in Text (27 days), WebDev (35 days) and Vision (10 days).

Embed

Credits LMArena.

Credits LMArena.

Credits Epoch AI.

Each frame links back to its page. Preview it.

Cite
Download
Post on X

Text: 27 days in first place

place 10 of 419 on 28 October 2026
1st10th
On the board
6 February 2026 to 28 October 2026, the newest snapshot
Score
1496 (1493–1499) on 28 October 2026 · 89,324 votes
First place
  • 19 February 2026 to 18 March 2026 · 27 days
Catch-up
No open-weights model has matched it in the 251 days since it first took first place.

WebDev: 35 days in first place

place 52 of 152 on 28 October 2026
1st10th
On the board
6 February 2026 to 28 October 2026, the newest snapshot
Score
1537 (1532–1542) on 28 October 2026 · 20,576 votes
First place
  • 6 February 2026 to 9 February 2026 · 3 days
  • 17 February 2026 to 4 March 2026 · 15 days
  • 16 March 2026 to 2 April 2026 · 17 days
Catch-up
Open weights matched it on 16 June 2026, 130 days after it first took first place: glm-5.2 (max).

Vision: 10 days in first place

place 11 of 167 on 19 October 2026
1st10th
On the board
26 March 2026 to 19 October 2026, the newest snapshot
Score
1312 (1306–1319) on 19 October 2026 · 26,841 votes
First place
  • 31 March 2026 to 10 April 2026 · 10 days
Catch-up
No open-weights model has matched it in the 202 days since it first took first place.

Epoch AI: Claude Opus 4.6

released 5 February 2026 · API access
  1. Chess Puzzles 21.0%
  2. EBR-bench 12.7%
  3. FrontierMath-2025-02-28-Private 40.7%
  4. FrontierMath-Tier-4-2025-07-01-Private 22.9%
  5. FrontierMath-Tier-4-v2-Private 26.8%
  6. FrontierMath-Tiers-1-3-v2-Private 66.0%
  7. Furniture Assembly 28.3%
  8. GPQA diamond 90.5%
  9. Mystery Game Puzzles 25.0%
  10. OTIS Mock AIME 2024-2025 94.4%
  11. SWE-Bench verified 78.7%
  12. SimpleQA Verified 47.0%

When each vendor retires it

every model’s dates

Sources

Arena scores come from people voting between two anonymous answers. First place is the highest score; the top two are often within each other’s margin of error. LMArena, Arena leaderboard dataset, revision 4491b06, retrieved 30 October 2026; Epoch AI, Capabilities & benchmarking, retrieved 30 October 2026. Both CC BY 4.0. Retirement dates: each vendor’s own deprecation page, quoted and kept, listed with the record.