Teamday

Chat & Reasoning Model · Anthropic

Claude Sonnet 5.5

Claude Sonnet 5.5 is Anthropic's newest Sonnet, released September 28, 2026. It succeeds Claude Sonnet 5 at the same price and with the same tokenizer, with a 1M-token context window and up to 128K output tokens. Anthropic reports it finishes agentic coding and multi-step tool-use work in far fewer model requests than Sonnet 5.

Use Claude Sonnet 5.5 in TeamDay →All models

At a glance (verified 2026-09-29)

Vendor
Anthropic
Type
Chat & Reasoning Model
Context window
1M tokens
Pricing
Base: $2 in / $0.20 cache hits / $10 out per 1M tokens. Cache writes: $2.50 5m / $4 1h.

Best for

Everyday codingAgents and tool useLong-context workCost-efficient daily driver

How to use Claude Sonnet 5.5 in TeamDay

Select Claude Sonnet 5.5 in TeamDay's chat model picker on the Claude Code harness, right below Claude Opus 5.5. Agents and chats previously set to Sonnet 5 run on it automatically.

Harness: claude-code →Max →Maya →

Frequently asked questions

How much does Claude Sonnet 5.5 cost per million tokens?

The same as Claude Sonnet 5: $2 per 1M input tokens, $0.20 for cache hits, and $10 per 1M output tokens; cache writes cost $2.50 (5-minute) or $4 (1-hour). It uses the same tokenizer, so token counts are unchanged.

When should I pick Claude Sonnet 5.5 over Opus 5.5?

Pick Sonnet 5.5 for everyday coding, agents, and long-context work at half Opus 5.5's price ($2 vs $4 per 1M input tokens). For the hardest long-horizon work, Anthropic points to an Opus model.

What changed from Claude Sonnet 5?

Effort levels were recalibrated, so TeamDay starts Sonnet 5.5 at medium effort — Anthropic's suggested level for agentic work. Thinking can no longer be switched off the old way, and forced tool use now returns an error, as on Claude Opus 5.5.

Is Claude Sonnet 5.5 available in TeamDay?

Yes — it is in TeamDay's chat model picker on the Claude Code harness, with reasoning effort selectable from low to max.

Related

Claude Sonnet 5.5is one of the models TeamDay's AI employees already run — no API keys, no infrastructure.

Hire your first AI employee →