Anthropic

Claude Opus 5

Jul 24, 2026ReasoningClosed weights

Compare

Intelligence

39.4

#74 of 751Rank 74 of 751 measured models

Coding

66.9

#65 of 298Rank 65 of 298 measured models

Output speed

53.9 t/s

#330 of 411Rank 330 of 411 measured models

Price/1M

$10

#633 of 677Rank 633 of 677 measured models

Context window

1M

#89 of 1256Rank 89 of 1256 measured models

Cost per task

$1.10

#141 of 200Rank 141 of 200 measured models

Claude Opus 5 (Low) scores 39.4 on the Artificial Analysis Intelligence Index, ranking 74th of 751 measured models. The closest models on this index are Muse Spark 1.2 (Xhigh) (39.6), DeepSeek V4.1 Flash (39.5), and Gemini 3.7 Flash (39.1). At $10 per million tokens (3:1 input/output blended price), it is more expensive than Muse Spark 1.2 (Xhigh) ($2), DeepSeek V4.1 Flash ($0.525), and Gemini 3.7 Flash ($1.50). With an output speed of 53.9 t/s, it is slower than Muse Spark 1.2 (Xhigh) (288 t/s), DeepSeek V4.1 Flash (209 t/s), and Gemini 3.7 Flash (383 t/s).