All times Eastern (ET)Provider pages checked daily at
ProSign inRSS

Claude Opus 4 (2025-05-14)

API ID claude-opus-4-20250514

claude-opus-4-20250514 was an Anthropic model, released May 2025.

Status
Retired
Released
May 2025
Shutdown
Jun 15, 2026

claude-opus-4-20250514 stopped serving on June 15, 2026. Recommended replacement: claude-opus-4-8.

How to migrate to Claude Opus 4.8

Benchmarks

Scores as each source published them
or

Comparing with Claude 3 Opus (2024-02-29), the previous model in this line. A score is compared only with the same benchmark from the same source, at the same setting where both have it. Full comparison: prices, limits, features

Arena

ArenaClaude Opus 4 (2025-05-14)Claude 3 Opus (2024-02-29)ChangeRankDateSource
Text
Other settings
  • 1414 ±4 vs 1322 ±3
By category
  • Math 1421 rank 128 vs 1312 (+109) at thinking 16k (vs no setting)
  • Creative writing 1433 rank 71 vs 1288 (+145) at thinking 16k (vs no setting)
  • Instruction following 1444 rank 74 vs 1314 (+130) at thinking 16k (vs no setting)
  • Hard prompts 1457 rank 106 vs 1330 (+127) at thinking 16k (vs no setting)
  • Long prompts 1466 rank 66 vs 1330 (+136) at thinking 16k (vs no setting)
  • Multi-turn 1439 rank 109 vs 1322 (+117) at thinking 16k (vs no setting)
  • Expert 1450 rank 118 vs 1319 (+131) at thinking 16k (vs no setting)
1426 ±41322 ±3+104120 of 415
35,909 votes		Oct 13, 2026source
Text: Coding
Other settings
  • 1464 ±7 vs 1355 ±6
1498 ±81355 ±6+14383 of 410
6,594 votes		Oct 13, 2026source
Vision
Other settings
  • 1188 ±13 vs 1061 ±10
1208 ±161061 ±10+14787 of 165
1,289 votes		Oct 9, 2026source
Search1152 ±6not rated35 of 42
31,506 votes		Oct 13, 2026source

Rated as claude-opus-4-20250514-thinking-16k; Claude 3 Opus (2024-02-29) as claude-3-opus-20240229. Some were rated at different settings; ratings on one arena share one scale, so the difference still reads directly. Ratings come from people voting between two anonymous answers; Text, Vision and Search are style-controlled, as on arena.ai. Data: Arena (arena.ai), CC BY 4.0.

Epoch AI

BenchmarkClaude Opus 4 (2025-05-14)Claude 3 Opus (2024-02-29)ChangeSettingRunSource
GPQA Diamond
Other settings
  • 69.2% vs 47.2%
76.3%47.2%+29.1 pts16k (vs no setting)May 22, 2025source
MATH Level 585.0%37.5%+47.5 ptsMay 22, 2025source
OTIS Mock AIME 2024-2025
Other settings
  • 60.0% vs not run 16k
  • 42.2% vs 4.7%
64.4%4.7%+59.7 pts27k (vs no setting)May 28, 2025source
SWE-bench Verified70.7%not runFeb 6, 2026source

Run independently by Epoch AI. Data: Epoch AI, Capabilities & benchmarking (epoch.ai/benchmarks), CC BY.

Price history

USD per 1M tokens, standard tier
$0.00$20.00$40.00$60.00$80.00$100.00Oct 2026In $15.00Out $75.00Oct 1, 2026 to Oct 15, 2026: Input $15.00, Output $75.00
$0.00$20.00$40.00$60.00$80.00$100.00Oct 2026In $15.00Out $75.00Oct 1, 2026 to Oct 15, 2026: Input $15.00, Output $75.00

We started recording this model's prices on Oct 1, 2026. The history grows with each daily check.

Show as a table
FromTypePrice
Oct 1, 2026Input$15.00
Oct 1, 2026Output$75.00

Prices

USD per 1M tokens, every type and band, by tier
TypePrice $/1MIn effect
Standard
input$15.00
cached input$1.50
cache write (5 min)$18.75
cache write (1 hour)$30.00
output$75.00
Batch
input$7.50
output$37.50

Changelog

Changes we detected on Anthropic's pages

No changes recorded yet.

Last checked · Sources: Anthropic models, Anthropic deprecations, Anthropic pricing, Anthropic model pages