Claude Opus 5
Senior Agentic Engineer at Anthropic | Long coding runs and computer use | 1M-token context
The basics
Checked Oct 19- Status
- Model ID
- Price
- Context
- Supported to
Updates
Strengths and weaknesses
Strengths
Weaknesses
- !
- !
- !
Role
Flagship model at Anthropic
Sep 2026 to now · 2 months
Jul 2026 to Sep 2026 · 2 months- Sep 22, 2026: Legacy model
- Jul 24, 2026: Flagship model
May 2026 to Jul 2026 · 2 months
Works with
From recorded changes
Agent · 5 shared changes
Agent · 2 shared changes
App · 1 shared change
API · 1 shared change
Agent · 1 shared change
Similar models
$5 in, $25 out · Compare
$4 in, $20 out · Compare
$2 in, $10 out
Cost vs peers
Tied for most expensive of 6 similar models by output price.
Standard API rates for uncached input on the shortest prompts, from each maker's pricing page, or from models.dev for models not researched here yet. Gemini 4 Argon: Introductory price through Nov 30, 2026, then $4 in and $20 out.
Cost and specs
Cost
Per 1M tokens
Specs
Where it works
Prices and uptime are OpenRouter's figures for each provider's cheapest endpoint. Uptime covers the last 24 hours.
Benchmarks
Higher is better. Green is Claude Opus 5. Dark marks are similar models.
Independent
Senior SWE-bench
2nd
Self-reported by Anthropic
Terminal-Bench 4.0
52.3%
OSWorld 2.1
74.0%
What people say
Reviews (1)
Lior GrossmanShipping software since 1993, now building with AII couldn't stand working with Opus 5. It was my daily model in Claude Code and Cowork from August until Opus 5.5 replaced it in late September 2026. It was intelligent and persistent, but it made more mistakes than Opus 4.8 and was less fun to work with. Its writing was verbose yet compressed, dense and elliptical, full of its own jargon and trying to be clever. I kept telling it to stop talking in codes and write like a human. It argued with my instructions and called unfinished tasks done. One scheduled task kept making false claims, so I moved it to Fable.
Oct 31, 2026
Incidents
Incidents per week, last 13 weeks. Most in one week: 1.
- Sep 22Anthropic reported elevated errors across several Claude models1 h 20 min
- Sep 3Anthropic reported elevated errors across several Claude models2 h 57 min
- Aug 24Anthropic reported elevated errors on Claude models3 h 24 min
- Aug 18Anthropic reported elevated errors on several Claude models2 h 41 min
- Aug 5Anthropic reported elevated errors on four Claude models7 h 9 min
Elevated errors, 6 times in 5 weeks
ResolvedGuessed passwords during testing
Maker-confirmedWriting habits
Creative writing · EQ-Bench, among 51 models tracked here
Slop score
6.6
Overused AI words and phrases. Less slop than 100% of tracked models.
Repetition
4.3
Repeated words and ideas. Less repetition than 26% of tracked models.
Story length
6,003
Average characters per story. Longer than 32% of tracked models.