By the toolist.ai editors. Researched from public sources, not hands-on tested by toolist.ai. Updated October 6, 2026; prices checked October 6, 2026.
Claude Opus 5.5: price, context window and benchmarks
Claude Opus 5.5 is Anthropic's current Opus model and the one Anthropic recommends starting with, at $4 per million input tokens and $20 per million output tokens, 20% below Claude Opus 5. It has the highest Artificial Analysis Intelligence Index of the six models we track: 57.6 at max effort.
Best for
- Agentic coding and long knowledge-work tasks where you want the top benchmark scores without Fable prices
- Replacing Claude Opus 5, at lower token prices and with higher scores on every shared benchmark
- Prompt-heavy agents, since cache reads cost $0.20 per million tokens
Key facts
| Fact | Claude Opus 5.5 | Source |
|---|---|---|
| Provider | Anthropic | Claude docs: Pricing |
| API model name | claude-opus-5-5 | Claude docs: Claude Opus 5.5 |
| Released | September 22, 2026 | Anthropic: Introducing Claude Opus 5.5 |
| Input price | $4 per million tokens | Claude docs: Pricing |
| Cached input price | $0.20 per million tokens | Claude docs: Pricing |
| Output price | $20 per million tokens | Claude docs: Pricing |
| Context window | 1,000,000 tokens | Claude docs: Claude Opus 5.5 |
| Maximum output | 128,000 tokens | Claude docs: Claude Opus 5.5 |
| Knowledge cutoff | June 2026 | Claude docs: Claude Opus 5.5 |
| Input and output | Text and images in; text out | Claude docs: Claude Opus 5.5 |
| Reasoning | Adaptive thinking is always on and cannot be turned off. Effort levels low to max; medium is the API default | Claude docs: Claude Opus 5.5 |
| API and cloud platforms | Claude API, Amazon Bedrock, Google Cloud and Microsoft Foundry | Claude docs: Claude Opus 5.5 |
| In the apps | Claude apps on Pro, Max, Team and Enterprise; not on Free | Claude plans and pricing |
| Data retention | Available under zero data retention arrangements; it is not one of the models that require 30-day retention | Claude docs: API and data retention |
Cache writes cost $5 per million tokens (5-minute cache) or $8 (1-hour cache). The Batch API costs half. There is no long-context surcharge. Fast mode costs $8 in and $40 out.
What Claude Opus 5.5 costs a month
Three typical workloads, priced at the standard API rates above. We calculated these; the provider does not publish them.
| Workload | Tokens a month | Cost a month |
|---|---|---|
| Support chatbot 100,000 replies, each 3,000 input and 400 output tokens, no caching | 300M in, 40M out | $2,000 |
| Document Q&A 5,000 questions over the same 60,000-token document set (90% cache hits), 800 output tokens each | 300M in, 4M out | $254 |
| Coding agent 200 tasks, each 1.5 million input tokens (90% cache hits) and 50,000 output tokens | 300M in, 10M out | $374 |
Reasoning models bill their thinking as output tokens, so heavy reasoning can raise the output count several times. Cache-write charges are not included.
Benchmarks
| Benchmark | Score | Source |
|---|---|---|
| Artificial Analysis Intelligence Index max effort | 57.6 | Artificial Analysis: Claude Opus 5.5 (max) |
| Intelligence Index at everyday effort medium effort, the API default | 51.2 | Artificial Analysis: Claude Opus 5.5 (medium) |
| Coding Agent Index Claude Code harness, max effort | 66.0 | Artificial Analysis: Coding agents |
| Terminal-Bench 4.0 max effort | 59.6% | Artificial Analysis: Claude Opus 5.5 (max) |
| Humanity's Last Exam max effort | 61.4% | Artificial Analysis: Claude Opus 5.5 (max) |
| GDPval-AA v2.1 (Elo) max effort | 1866 | Artificial Analysis: Claude Opus 5.5 (max) |
| SciCode max effort | 66.9% | Artificial Analysis: Claude Opus 5.5 (max) |
| Terminal-Bench-Science 0.1 max effort | 59.0% | Artificial Analysis: Claude Opus 5.5 (max) |
Speed: about 79.2 output tokens a second, 20.6 seconds to the first token (medium effort, Anthropic API; Artificial Analysis: Claude Opus 5.5 (medium)).
Benchmark and speed data marked as such come from Artificial Analysis.
What changed
Compared with Claude Opus 5, token prices fall from $5 and $25 to $4 and $20 per million, cache reads from $0.50 to $0.20, and the knowledge cutoff moves from May to June 2026. Thinking can no longer be turned off, and the default effort drops from high to medium. (Claude docs: What's new in Claude Opus 5.5)
Claude Opus 5.5 compared
Other models
- GPT-6.1 Sol: $2 in, $10 out per million tokens
- GPT-6 Sol: $2 in, $10 out per million tokens
- GPT-6 Astra: $10 in, $50 out per million tokens
- Claude Fable 5.1: $10 in, $50 out per million tokens
- Claude Opus 5: $5 in, $25 out per million tokens
All prices side by side: AI model API price table. How we research: our method.
Sources
- Anthropic: Introducing Claude Opus 5.5 (checked 2026-10-06)
- Claude docs: Claude Opus 5.5 (checked 2026-10-06)
- Claude docs: What's new in Claude Opus 5.5 (checked 2026-10-06)
- Claude docs: Pricing (checked 2026-10-06)
- Claude plans and pricing (checked 2026-10-06)
- Claude docs: API and data retention (checked 2026-10-06)
- Artificial Analysis: Claude Opus 5.5 (max) (checked 2026-10-06)
- Artificial Analysis: Claude Opus 5.5 (medium) (checked 2026-10-06)
- Artificial Analysis: Coding agents (checked 2026-10-06)
- OpenRouter: Claude Opus 5.5 (checked 2026-10-06)