Claude Opus 5.5
Anthropic's new flagship: Fable 5.1-level coding and analytics, but faster and more affordable
included with Max plan
- Context
- 1,000,000 tokens
- Output
- up to 128,000 tokens
- Input
- Text, images
- Type
- Flagship, always with reasoning
- Developer
- Anthropic
- Access
- ChatPlus PRO
Claude Opus 5.5 is the first model in the new Claude 5.5 family, introduced on September 22, 2026. It performs at the level of Claude Fable 5.1 on most tasks, while outperforming it in Anthropic benchmarks for agentic programming, document work, and computer use. It generates text 30% faster than Opus 5 and costs 40% less on average. Its writing style has also been redesigned: key points come first, with less jargon and filler. Context window: up to 1M tokens. Available on ChatPlus with the Max plan.
Opus 5.5 is a rare case where a new version is simultaneously smarter, faster, and cheaper than its predecessor. Anthropic positions it as a Fable 5.1-level model without the flagship price tag: input tokens are 20% cheaper, and because the model takes fewer steps per task, total savings can reach 40%.
It excels most at long-form engineering work. Early testers report migrating 680 thousand lines of code in less than a day and auditing a 200 thousand-line project in three hours—where Opus 5 took more than twenty. In Terminal-Bench 4.0, the model scored 66.4%—higher than Fable 5.1 and GPT-6 Astra.
The second notable change is its writing style. Opus 5 was often criticized for being verbose, but 5.5 writes differently: conclusion first, then details, without bureaucratic phrasing, so its drafts need fewer edits. On ChatPlus, Claude Opus 5.5 is available on the Max plan.
What it's good at
Agentic coding
Long migrations and codebase audits: in an Anthropic test, it rewrote HAProxy from C to Rust in 9.5 hours versus 12 for Fable 5.1.
Documents and analytics
Reports, financial models, presentations. Scored 1846 Elo on GDPval-AA, above Fable 5.1 and Opus 5.
Clear writing
Puts the key point first, writes without jargon, and follows specified style guidelines.
Efficiency
Uses fewer steps and tokens per task — around 40% cheaper than Opus 5 with higher quality.
Claude Opus 5.5 and every other model — one plan, one window.
Where it's strongest
Claude next to the others
How Claude Opus 5.5 and the flagships of other labs look side by side. They're all in one ChatPlus subscription.
| Claude | DeepSeek | GPT | Gemini | Grok | GLM | Kimi | Qwen | Muse | GLM 5.3 Flash | |
|---|---|---|---|---|---|---|---|---|---|---|
| Developer | Anthropic | DeepSeek | OpenAI | xAI | Z.ai | Moonshot | Alibaba | Meta | Z.ai | |
| Model | Claude Opus 5.5 | DeepSeek V4.1 Reasoner | GPT-6 Astra | Gemini 3 Pro | Grok 4 | GLM 5 | Kimi K3 | Qwen 3.8 Max | Muse Spark 1.3 | GLM 5.3 Flash |
| Context | 1M | 128K | 1M | 1M | 256K | 1M | 256K | 256K | 1M | 1M |
| Understands images | — | — | ||||||||
| Live web data | — | — | — | — | — | — | — | |||
| Open source | — | — | — | — | — | |||||
| Key strength | Code | Reasoning | Computer use | Multimodality | Timeliness | Code and agents | Long context | Complex tasks | Agents and code | Code and agents |
| On ChatPlus |