Back to models

Claude Opus 4.5

ECI#45
150
Anthropic
Nov. 24, 2025
Closed weights

Claude Opus 4.5 is a closed-weights model from Anthropic, released November 24, 2025. It has an ECI score of 150 and ranks 45th of 249 tracked models. At the time of its release, it ranked 3rd of 173 tracked models. At $5/Mtok for input tokens and $25/Mtok for output tokens, it is more expensive per token than similarly-capable models.

Input price$5 / 1M tokens
Output price$25 / 1M tokens
ParametersUnknown
Training computeUnknown

Claude Opus 4.5 vs select alternatives

DomainBenchmark
Claude Opus 4.5
Anthropic
Nov. 24, 2025
Closed weights
Anthropic
Sep. 29, 2025
Closed weights
Google DeepMind
Nov. 18, 2025
Closed weights
OpenAI
Oct. 7, 2025
Closed weights
ECI scoreBest score 166
150147153150
Ranking#45#61#35#43
Games
Chess PuzzlesBest score 72.0%
12.0%
12.0%
31.0%
n/a
14.3%
2.4%
n/a
n/a
Mystery Game PuzzlesBest score 84.0%
22.0%
17.0%
n/a
n/a
Mathematics
34.4%
23.9%
n/a
55.8%
FrontierMath Tier 4 (v2)Best score 97.6%
4.9%
2.4%
n/a
19.5%
OTIS Mock AIME 2024-2025Best score 100.0%
86.1%
77.8%
91.4%
n/a
Science
GPQA DiamondBest score 95.8%
86.0%
82.3%
92.6%
n/a
Software engineering
SWE-bench VerifiedBest score 83.5%
76.7%
71.3%
72.9%
n/a
World knowledge
SimpleQA VerifiedBest score 75.6%
45.7%
30.7%
n/a
n/a

Model metadata