- Intelligence
- 50.8
- Rank 18 of 751 measured models
- Coding
- 78.0
- Rank 5 of 298 measured models
- Output speed
- 54.1 t/s
- Rank 327 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
- $5.86
- Rank 196 of 200 measured models
Claude Opus 5 (Max) scores 50.8 on the Artificial Analysis Intelligence Index, ranking 18th of 751 measured models. The closest models on this index are Muse Wonder (Max) (52.4), GPT-6.1 Sol (51.8), and Claude Fable 5 (batch) (49.6). At $10 per million tokens (3:1 input/output blended price), it is more expensive than Muse Wonder (Max) ($8) and GPT-6.1 Sol ($4). With an output speed of 54.1 t/s, it is about as fast as GPT-6.1 Sol (54.3 t/s), slower than Muse Wonder (Max) (145 t/s).
Additional benchmarks16
Design Arena · 3d59.7%
Design Arena · Agenticgamedev53.9%
Design Arena · Androidnative55.9%
Design Arena · Asciiart67.4%
Design Arena · Blender3d54.9%
Design Arena · Codecategories57.3%
Design Arena · Dataviz58.3%
Design Arena · Fullstack57.3%
Design Arena · Gamedev58.3%
Design Arena · Godotgamedev54.2%
Design Arena · Mobileapps67.1%
Design Arena · Python Pptxslides54.5%
Design Arena · Svg58.2%
Design Arena · Uicomponent60.8%
Design Arena · Webapps54.3%
Design Arena · Website55.9%
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.
- Gemini 4 Argon (High)
Google · $4 · 81.5 t/s
52.6+1.8$481.5 t/sCompare - Muse Wonder (Max)
Meta · $8 · 145 t/s
52.4+1.6$8145 t/sCompare - GPT-6.1 Sol
OpenAI · $4 · 54.3 t/s
51.8+1.0$454.3 t/sCompare - Claude Fable 5 (batch)
Anthropic · — · —
49.6−1.2——Compare
