Claude Opus 4.8
Rank of 27196from #13
Known for
Codebase-scale migrations and long agentic sessions that need a model that checks its own work.
The last Opus before Opus 5, and the one that brought dynamic workflows to Claude Code: plan, fan out to hundreds of subagents, verify, return. Still sold, still sharp on agentic judgment.
What it does7
- 1M-token context, 128K output
- Four times less likely than Opus 4.7 to let its own code flaws pass
- Effort control on claude.ai and Cowork
- Mid-task system entries without breaking the cache
- Fast mode at 2.5x speed
- Better PDF and diagram reasoning
- Legacy status, retirement not before May 2027
From Claude Opus 4.8’s own site. Not used for ranking.
41index score · LLMs
How well does it score?
Compared with everyone else on this board.
—votes · crowd
Reading votes…
Votes don’t change the score.