Anthropic: Claude Sonnet 5.5

Fast Claude model for everyday coding, agents, and knowledge work

claude-sonnet-5-5

AvailableAnthropicChatCached input 95% less

Input / Output-40% ~ -70%: $2.00 / $10.00$0.60 / $3.00
Context: 1M
Released: Sep 28, 2026
Max output: 128K
Modalities: VisionChat
Capabilities: Tool usePrompt CacheReasoning

About Claude Sonnet 5.5

Claude Sonnet 5.5 is Anthropic's current Sonnet model, released on September 28, 2026, which it describes as the best combination of speed and intelligence. It generates output faster than Sonnet 5 and needs fewer tokens per task. It handles everyday coding, agents, and computer use, with a knowledge cutoff of June 2026.

Where it works well

Anthropic reports a very large jump over Sonnet 5 on Terminal-Bench 4.0, its agentic terminal-coding benchmark.
Anthropic reports strong OSWorld 2.1 results, and says it is the first Sonnet to finish Pokemon Red from screenshots alone.
Batches tool calls better, so tasks finish in fewer steps and cost per task falls, per Anthropic.
Thinking can be reduced with the between_tools setting, which turns off up-front thinking for faster replies.

When to choose another model

Anthropic says Opus 5.5 is clearly stronger on complex, open-ended work that needs sustained judgment.
New safeguards restrict some cybersecurity tasks, and higher-risk requests fall back to Sonnet 5.
Forced tool use returns an error, and non-default temperature, top_p, or top_k values are rejected.

Getting started

Create API key
Create a key in Console, then use it with every model on the platform.

Send your first request

Copy the example for your language and run it against the endpoint.

Text to textMessages API

POST/v1/messages

API format:

curl https://api.tokenlab.sh/v1/messages \
  -H "Content-Type: application/json" \
  -H "x-api-key: sk-xxx" \
  -H "anthropic-version: 2023-06-01" \
  -d '{
    "model": "claude-sonnet-5-5",
    "max_tokens": 1024,
    "messages": [
      {"role": "user", "content": "Hello!"}
    ]
  }'

Pricing

The Verified price applies to Verified, which costs less on most models. The Official price is the model maker's published price and applies to the more reliable Official route. Auto bills the route that completes the request.

Default spec

per 1M tokens

Official price: Input $2.00 / Output $10.00 / Cache read $0.10 / Cache write $2.50
Verified price: Input $0.60 / Output $3.00 / Cache read $0.06 / Cache write $0.75
Discount: -40% ~ -70%

Prompt cache pricing

Cache read

Official price: $0.10
Verified price: $0.06
Discount: -40%

Cache write

Official price: $2.50
Verified price: $0.75
Discount: -70%

Tool fees

Charged per call, only when the model uses the tool.

Web search

Official price: $0.01/search
Verified price: $0.003/search
Discount: -70%

	Official priceper 1M tokens	Verified priceper 1M tokens	Discount
Default spec	Input $2.00 / Output $10.00 / Cache read $0.10 / Cache write $2.50	Input $0.60 / Output $3.00 / Cache read $0.06 / Cache write $0.75	-40% ~ -70%
Prompt cache pricing
Cache read	$0.10	$0.06	-40%
Cache write	$2.50	$0.75	-70%
Tool feesCharged per call, only when the model uses the tool.
Web search	$0.01/search	$0.003/search	-70%

Usage & activity

Success rate is the share of requests that completed. Latency is how long a full response takes; P95 means 95% of requests finished within that time.

Usage & availability

Last 24 hours

Requests
Success rate
P95 latency
Total tokens

Model performance

Last updated Oct 16, 2026, 11:53 AM UTC

30-day success rate: 97.0%
7-day median latency: 4.5sn=10,833
7-day P95 latency: 19.1sn=10,833
30-day requests: 20K+
Last active: 50 seconds ago

Data is based on aggregate user requests, excluding status checks.

Open in Console

Open Claude Sonnet 5.5 in Console with a prompt ready to edit or send.

Help me try claude-sonnet-5-5 with a short message at /v1/messages. Show the reply, latency, and cost.

Use cases

Best for

Reasoning
Vision

Interactive coding assistant
Power an editor or terminal agent where the developer waits on each step, and fast replies and fewer tool round-trips matter.
Computer-use workflows
Operate web apps and desktop tools from screenshots to file forms, move data between systems, and confirm results.
Knowledge-work assistants
Draft documents, analyze spreadsheets, and answer questions across long files, with Anthropic reporting knowledge-work scores close to Opus 5.5.

Prompt examples

Fix the failing tests in this branch, then run the linter and tell me what you changed in each file.

Open the expense tool, submit the receipts in this folder, and list any that were rejected.

Read this 80-page report and give me a two-paragraph brief plus the five numbers a CFO should check.

This model has conditional pricing. Monthly token totals alone cannot produce a reliable estimate; use the detailed pricing for the request specification, cache, and applicable time window.

FAQ

What is Claude Sonnet 5.5 best at?

Fast everyday coding, agents, and computer use. Anthropic markets it as the best mix of speed and intelligence, and reports large agentic coding gains over Sonnet 5 along with knowledge-work scores near Opus 5.5.

How is Sonnet 5.5 different from Opus 5.5?

Sonnet 5.5 is faster and smaller, and its thinking can be reduced with between_tools. Opus 5.5 always thinks, defaults to medium effort, and Anthropic calls it clearly stronger for complex open-ended work needing sustained judgment.

Can I turn off thinking on Sonnet 5.5?

Partly. Adaptive thinking is on by default, and the lowest setting, between_tools, turns off up-front thinking while keeping thinking between tool calls. It works at high effort or below. Forced tool use is not supported.

What breaks when moving from Sonnet 5 to 5.5?

Anthropic lists forced tool use returning an error, thinking blocks tied to the model, the older computer_20251124 tool being rejected, advisor pairings changing, and text between tool calls arriving in thinking blocks. Test streaming UIs before switching.

How much does Claude Sonnet 5.5 cost?

On TokenLab, Claude Sonnet 5.5 costs Input $0.60 / Output $3.00 per 1M tokens. The pricing table above shows the full breakdown. Rates depend on the billing unit, specification, and usage. Compare matching conditions in the model's detailed pricing; a single rate does not determine the total cost.

What are the context window and output limit of Claude Sonnet 5.5?

Claude Sonnet 5.5 accepts up to 1,000,000 tokens of context and returns up to 128,000 tokens in one response.

Which endpoint should Claude Sonnet 5.5 use?

Use https://api.tokenlab.sh/v1/messages for Claude Sonnet 5.5. The request example below shows the matching code shape.

Which operations does Claude Sonnet 5.5 support?

Claude Sonnet 5.5 supports Text to text. Select an operation above to see its endpoint and request example.