Anthropic
Claude Opus 5.5
Anthropic's September 22, 2026 model for complex coding and agentic enterprise work, the first Opus to get cheaper: $4 input and $20 output per million tokens, down 20% from Opus 5. Thinking is always on with a default medium effort. Provides a 1M-token context window and up to 128K output (300K on the Batch beta), with cache reads at $0.20 and a 50% Batch discount; an optional Fast serving mode runs at $8/$40 per million tokens. On Artificial Analysis it leads the Intelligence Index at 58 (currently #1 of 212 models) and posts Terminal-Bench 4.0 of 60%, while Anthropic separately claims 66.4% at extra-high effort.
- Provider
- Anthropic
- Context
- 1MCatalog · As of Sep 25, 2026Under 1d old
- yno subscription tier
- Max
- Released
- Sep 22, 2026
- Speed
- Medium
- Reasoning
- Expert
- Modality
- Multimodal
Benchmarks
Quality
- AA Intelligence Index
- GPQA Diamond
- 94%Artificial Analysis · As of Sep 25, 2026Under 1d old
Speed and latency
- Output speed
API pricing
Not available
Indexes use points; evaluations use accuracy percentages. API measurements do not measure yno application performance. Methodology
Capabilities
- Reasoning
- Code
- Agentic
- Vision
- Computer Use
- Long Horizon
Use cases
- Agentic software engineering
- Workflow automation
- Scientific research
- Long-running agents
Strengths
- AA Intelligence Index 58 (#1 at launch)
- 1M context / 128K output (300K Batch beta)
- First Opus with a price cut: $4/$20 per MTok
- Thinking always on, adjustable effort
Best for
Teams that want the top AA-ranked reasoning model for complex coding and enterprise agents
No credit card required
Related models
Frequently asked
- What is Claude Opus 5.5?
- Anthropic's September 22, 2026 model for complex coding and agentic enterprise work, the first Opus to get cheaper: $4 input and $20 output per million tokens, down 20% from Opus 5. Thinking is always on with a default medium effort. Provides a 1M-token context window and up to 128K output (300K on the Batch beta), with cache reads at $0.20 and a 50% Batch discount; an optional Fast serving mode runs at $8/$40 per million tokens. On Artificial Analysis it leads the Intelligence Index at 58 (currently #1 of 212 models) and posts Terminal-Bench 4.0 of 60%, while Anthropic separately claims 66.4% at extra-high effort.
- How much does Claude Opus 5.5 cost in yno.ai?
- Claude Opus 5.5 is available on the Max tier of yno.ai.
- What can Claude Opus 5.5 do?
- Claude Opus 5.5 is best for Teams that want the top AA-ranked reasoning model for complex coding and enterprise agents. Its main capabilities include Reasoning, Code, Agentic, Vision, Computer Use, Long Horizon.
- How does Claude Opus 5.5 compare to other models?
- Claude Opus 5.5 excels at AA Intelligence Index 58 (#1 at launch) and is recommended for Agentic software engineering, Workflow automation, Scientific research, Long-running agents. See related models below.
- How do I use Claude Opus 5.5 in yno.ai?
- Sign up for yno.ai, select Claude Opus 5.5 from the model picker, and start chatting.