Claude Opus 5.5 Is Now Available

Anthropic released Claude Opus 5.5 today. It is now the recommended Anthropic model for demanding work in AI Data Stream — the most capable option at standard pricing, one tier below the premium Fable 5.1. Opus 5 remains available as the previous generation, so you can compare the two on your own analytics before switching.

Pricing

This is the rare release that is both better and cheaper. On the standard API we use, Anthropic prices Opus 5.5 below Opus 5:

• Input: $4 per million tokens (Opus 5: $5)
• Output: $20 per million tokens (Opus 5: $25)
• Cache reads: $0.20 per million tokens (Opus 5: $0.50)

The cache-read cut matters more than it looks. Every turn of a conversation in AI Data Stream re-reads the system prompt, the tool definitions and the earlier turns from Anthropic’s prompt cache, so on long analysis sessions cache reads are a large share of the bill. Anthropic puts the overall saving at around 40% compared with Opus 5 for typical agentic workloads.

What’s New

Anthropic positions Opus 5.5 as performing at the level of Fable 5.1 — the premium model that costs $10/$50 per million tokens — while running more than 30% faster than Opus 5. The gains that matter most for analytics work are in long, multi-step tasks: on GDPval-AA, Artificial Analysis’s benchmark of real professional work across 44 occupations, Opus 5.5 scores 1846 Elo against 1708 for Opus 5 (and 1735 for Fable 5.1), and on Zapier’s AutomationBench, which tests real business workflows across connected apps, it passes 40.0% against 26.9% for Opus 5. In practice that shows up as fewer wasted tool calls and more reliable follow-through when you ask it to pull data from several connections and reconcile them. See Anthropic’s announcement for the full details.

Opus 4.8 Has Been Retired

We have removed Opus 4.8 from the model picker. If it was your default, you will move to Opus 5.5 automatically the next time you start a conversation — no action needed. Opus 5 is still available if you prefer to stay on the previous generation.

Effort Levels

Opus 5.5 supports the full effort scale: Low, Medium, High, Extra High, Max. Note that Anthropic’s own default for this model is Medium, one step down from Opus 5’s High — the picker still starts on the effort you saved for your Anthropic provider, so nothing changes unless you change it. If you had been running Opus 5 at High mainly to keep costs in check, it is worth trying Opus 5.5 at Medium and Extra High on the same question: the price drop gives you room to spend more effort for the same money.

The Full Claude Lineup

• Claude Fable 5.1 — Most powerful, premium; for the hardest reasoning
• Claude Fable 5 — Previous-generation Fable
• Claude Opus 5.5 — Latest frontier model, for demanding tasks
• Claude Opus 5 — Previous-generation Opus
• Claude Sonnet 5 — Best balance of speed and capability (default)
• Claude Sonnet 4.6 — Previous generation
• Claude Haiku 4.5 — Fastest, great for quick queries

Switching Models

1. Go to Settings > AI Providers
2. Click your Anthropic provider
3. Select your preferred model from the dropdown
4. Save

Or select it from the model picker when starting a new conversation.

Questions?

If you have any questions, contact us or check our AI providers documentation.