Anthropic
Claude Opus 5
ReasoningClosed weights
- Intelligence
- 44.8
- Rank 39 of 751 measured models
- Coding
- 74.3
- Rank 32 of 298 measured models
- Output speed
- 55.5 t/s
- Rank 328 of 411 measured models
- Price/1M
- $10
- Rank 633 of 677 measured models
- Context window
- 1M
- Rank 89 of 1256 measured models
- Cost per task
- $2.19
- Rank 174 of 200 measured models
Claude Opus 5 (Medium) scores 44.8 on the Artificial Analysis Intelligence Index, ranking 39th of 751 measured models. The closest models on this index are DeepSeek V4.1 Flash 1017 (Max) (45.1), GLM-5.3 (44.8), and Grok 4.6 (44.3). At $10 per million tokens (3:1 input/output blended price), it is more expensive than DeepSeek V4.1 Flash 1017 (Max) ($0.525), GLM-5.3 ($2.15), and Grok 4.6 ($3). With an output speed of 55.5 t/s, it is slower than DeepSeek V4.1 Flash 1017 (Max) (204 t/s), GLM-5.3 (77.7 t/s), and Grok 4.6 (80.3 t/s).
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- Qwen3.8 Max (0902)
Alibaba · $3 · 41.7 t/s
45.4+0.6$341.7 t/sCompare - DeepSeek V4.1 Flash 1017 (Max)
DeepSeek · $0.525 · 204 t/s
45.1+0.3$0.525204 t/sCompare - GLM-5.3
Z.AI · $2.15 · 77.7 t/s
44.8$2.1577.7 t/sCompare - Grok 4.6
SpaceXSI · $3 · 80.3 t/s
44.3−0.5$380.3 t/sCompare
