Claude Opus 5.5 Released: Near-Fable Performance at a Lower Price
Anthropic released Claude Opus 5.5 on 22 September 2026, the first model in the Claude 5.5 family. It's priced at $4/$20 per million tokens (down from $5/$25 for Opus 5), has a 1M-token context window and 128k output, always-on adaptive thinking, and is now the default model in Claude Code.
Anthropic released Claude Opus 5.5 on 22 September 2026, the first model in its new Claude 5.5 family. Anthropic says it performs at the level of Claude Fable 5.1 on most tasks while costing about 40% less to run than Opus 5.
The details
| Opus 5.5 | |
|---|---|
| Model ID | claude-opus-5-5 |
| Price (per million tokens) | $4 input / $20 output (Opus 5: $5 / $25) |
| Cache reads | $0.20 per million tokens — 0.05x input, half the usual ratio |
| Context window | 1M tokens by default |
| Max output | 128k tokens |
| Thinking | Adaptive thinking, always on |
| Batch API | $2 / $10 |
A few behaviours changed for API users:
- Thinking can't be turned off, and
tool_choiceof typeanyortoolreturns a 400 error. - Fast mode is available as a research preview on the Claude API at $8 / $40 per million tokens.
In Claude Code
On the Anthropic API and subscription plans, the opus alias — and the default model for Pro, Max, Team and Enterprise — now resolve to Opus 5.5. Its default effort level is medium; raise it with /effort for hard problems. (Effort Levels and Ultrathink, How to Change the Model)
Fast mode in Claude Code now defaults to Opus 5.5 too, billed from usage credits on subscription plans. (Claude Code Fast Mode)
What it means for you
- API users on Opus 5: switching the model ID is a straight 20% price cut on input and output, plus much cheaper cache reads. Test your prompts first — remove any code that disables thinking or forces tool choice.
- Claude Code users: you're probably already on it. Check with
/model. - Cost-sensitive builders: Sonnet 5.5 (released a week later) and Haiku 5.5 are cheaper again for routine work. (Opus vs Sonnet vs Haiku)
Source: Claude Platform release notes · Pricing
EasySpawn runs Claude Code on a persistent server with your own Claude plan or API key — so new models like Opus 5.5 are available the day they ship, with no markup. See how it works or join the waitlist.
Related: Claude Sonnet 5.5 Released · Claude Haiku 5.5 Released · Claude API Pricing Explained · What Is a Reasoning Model?
Keep reading
Claude Fable 5.1 and Mythos 5.1: Anthropic's Top Models Get an Update
On 1 September 2026 Anthropic released Claude Fable 5.1 (generally available) and Claude Mythos 5.1 (limited availability). Fable 5.1 costs $10/$50 per million tokens with unusually cheap $0.25 cache reads, a 1M-token context and 128k output. What it's for, what's restricted, and how to use it in Claude Code.
Gemini API: New Antigravity Agent Version, Gemini 3.8 Flash GA and Model Deprecations
Google's Gemini API replaced antigravity-preview-05-2026 with antigravity-preview-09-2026 on 17 September (the old version shut down on 5 October), made Gemini 3.8 Flash generally available on 2 September, limited Gemini 2.5 access to recent users, and deprecated gemini-3.1-flash-image. What developers need to change.