Anthropic · Claude 4
Claude Opus 4.8
Older Opus model (May 2026) kept available as a legacy option and safeguard fallback.
- AA Intelligence
- 41.8#18 of 142
- Input / output per 1M
- $5.00/ $25.00
- Output speed
- 56tok/s
- Context window
- 1M tokens
Composite indexes
- Artificial Analysis Intelligence Index#18 of 14242
Leader: Claude Opus 5.5 at 58
Coding
- SciCode#24 of 9454.4%
Leader: Claude Opus 5.5 at 66.9%
- Terminal-Bench 4.0#20 of 8823.6%21.7%
Leader: Claude Mythos 5.1 at 60.9%
- Terminal-Bench 2.x#17 of 10384.6%
Leader: Claude Fable 5.1 at 91.4%
Agents and tool use
Reasoning and knowledge
- GPQA Diamond#26 of 14292%
Leader: GPT-6 Astra at 96.1%
- Humanity’s Last Exam#10 of 14748.7%
Leader: Claude Opus 5.5 at 61.4%
- CritPt#12 of 13920.9%
Leader: GPT-5.6 Sol at 32.3%
- ARC-AGI-2#11 of 2572.1%
Leader: GPT-6 Astra at 95%
- ARC-AGI-3#6 of 91.5%
Leader: GPT-6 Astra at 62.7%
Vision and long context
- AA Long Context Reasoning#49 of 13977.7%
Leader: Kimi K3 at 88.7%
Human preference
- LMArena Text#13 of 541481
Leader: Claude Fable 5 at 1506
- LMArena WebDev#26 of 541555
Leader: Claude Opus 5.5 at 1818