Access Anthropic Claude Sonnet 5.5—also searched as Sonnet 5.5— through EvoLink's unified chat API. Test long-context coding, tool use, prompt caching before integrating.
Anthropic·Text Generation·Available
From $1.900 / 1M input tokens$2.000 official price-5%
Adaptive thinkingEffort controlTool usePrompt cachingChat + Messages API
Production routeThis rate reflects platform-side availability — only confirmed server errors (HTTP 500 / empty response) count as failures. User-side issues (content moderation, invalid params, cancellation) plus rate limits, timeouts and auth errors are excluded. Before real traffic arrives, empty buckets may display as available.Live
Live
Context
1M context · 128K max output
Best For
Everyday agentic coding, tool-using agents, high-volume production traffic
Anthropic’s newest Sonnet model for everyday agentic coding, tool-using agents, and high-volume production traffic, at the same list price as Claude Sonnet 5. Five billing dimensions — input, cache write, cache read, output, and the web search tool — are metered separately.
Claude Sonnet 5.5
Anthropic’s newest Sonnet model
Selected
Model ID
claude-sonnet-5-5
Best for
Fixing failing tests and multi-file changes in a real repository, pull-request review, agents that call search and internal tools, and computer-use workflows — the everyday work where a Sonnet-rate route should finish the task without escalating to Opus.
Input
$1.900 / 1M-5%
129.2 cr / 1M$2.000official price
Cache write
$2.375 / 1M-5%
161.5 cr / 1M$2.500official price
Cache read
$0.192 / 1M-4%
13 cr / 1M$0.200official price
Output
$9.500 / 1M-5%
646 cr / 1M$10.000official price
All rates are per 1M tokens, shown in USD and credits, and reflect your account's current pricing.
Claude Sonnet 5.5 pricing
Estimate what one Claude Sonnet 5.5 request costs before you integrate. The calculator uses your account's current rates, with official pricing as a reference.
Request calculator
Enter the token mix for one request and the number of successful tool calls.
Web search depends on the selected route and is unavailable on some routes. Confirm support before sending web_search; the rates below apply only where it is supported.
Web search$0.010/ call·0.68 cr / call
Claude Sonnet API access with Sonnet 5.5
Use Claude Sonnet API through EvoLink with Claude Sonnet 5.5, the current Sonnet version on this page. Connect coding agents and tool-based workflows through EvoLink’s unified API, with a 1M-token context window and up to 128K output tokens. Compare $1.900 input / $9.500 output per million tokens, reuse one API key across models, and check thinking and tool-choice settings before your first request.
Claude Sonnet 5.5 is served on EvoLink under the model ID claude-sonnet-5-5 through Chat Completions · Anthropic Messages, with the same API key and balance you use for every other model. It offers a 1M context window and up to 128K output tokens, plus long-context coding, tool use, prompt caching.
Claude Sonnet 5.5 specs and capabilities
Numbers come from the EvoLink route configuration; capabilities are what the API exposes today.
Context window
1M tokens
Max output
128K tokens
Input
Text + images
Output
Text · JSON (structured output) · tool calls
Reasoning
Thinking mode (on by default)
Tool use
Function calling with multi-step tool sequences
Prompt caching
Cache write + cache read rates
Protocols
Chat Completions · Anthropic Messages
Model ID
claude-sonnet-5-5
Where Claude Sonnet 5.5 is the right default route
Sonnet 5.5 is Anthropic’s everyday Claude for coding, agents, and enterprise work. Route to it by the cost of a completed task, then tune effort before touching the prompt.
Everyday agentic coding in a real repository
Fix failing tests, carry changes across files, and review pull requests. Anthropic reports that on most agentic coding evaluations Sonnet 5.5 at medium effort scored higher than Sonnet 5 at high effort, in far fewer model requests. Measure accepted patches, not snippet quality.
Agents that depend on connected tools
Support, research, and account agents that call search, internal APIs, or skills. Anthropic reports more reliable use of connected tools than Sonnet 5 and steadier adherence to the system-prompt role. If your prompts tell the model to avoid tools, remove that language before you evaluate.
Computer use and screen-driven workflows
Operate interfaces and read screenshots step by step. Anthropic reports that at high effort Sonnet 5.5 completed substantially more computer-use tasks than Sonnet 5 at its highest effort, with under a third of the tokens. On the Claude API it accepts computer use only through the computer toolset.
When another route is the better choice
The hardest long-horizon reasoning and analyst-grade deliverables belong on Claude Opus 5.5, with Claude Fable 5.1 as the escalation above it. Keep Claude Sonnet 5 where a workload is already validated on it and you have not re-tested yet.
Two ways to use Claude Sonnet 5.5: EvoLink API or Agent
Use the EvoLink API for product backends and batch jobs, or call Claude Sonnet 5.5 from Codex, Claude, or Gemini for coding and analysis workflows. Both paths share the same EvoLink API key, balance, model ID, and request history.
Option 1
Integrate with the EvoLink API
Best for: product backends, batch jobs, automated pipelines
Send OpenAI-compatible Chat Completions (or Anthropic Messages) requests to EvoLink and control the model ID, system prompt, output budget, tools, and structured output.
1Create an EvoLink API key in the console
2Point your OpenAI or Anthropic SDK at the EvoLink base URL and select the model ID shown above
3Send one representative request and read the usage field for input, cached, and output tokens
4Set max_tokens and retries per task; keep tool-call IDs and results across turns
Best for: coding, review, and analysis tasks in Codex, Claude, and Gemini
Give the Agent the task, the inputs to include, and the acceptance criteria. It assembles the request, calls Claude Sonnet 5.5 through EvoLink, and returns the answer with token usage.
1Set EVOLINK_API_KEY in your local environment; never put it in code or a prompt
2Describe the task, the inputs to include, and the expected output format
3Ask the Agent to call Claude Sonnet 5.5 through EvoLink and show the request before sending
4Let the Agent report the answer, token usage, and any error body
Claude Sonnet 5.5 API code example and error handling
This example shows the shortest runnable request: an OpenAI-compatible Chat Completions call with a system prompt, a user message, and an output budget. Open the API tab for the complete parameter and response reference.
curl -X POST https://direct.evolink.ai/v1/chat/completions \
-H "Authorization: Bearer $EVOLINK_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "claude-sonnet-5-5",
"messages": [
{ "role": "system", "content": "You are a coding agent working in this repository." },
{ "role": "user", "content": "Fix the failing test and explain the change:\n<test output>" }
],
"max_tokens": 16000
}'
# Anthropic Messages is also available at /v1/messages with the same model
# ID.
# Thinking is on by default and counts toward max_tokens, so leave room for
# it. The response includes choices[0].message and a usage object
# (prompt_tokens, completion_tokens, and cache read / write tokens when
# caching applies).
Invalid request or unsupported parameter
Check the model ID, the messages array, and parameter ranges against the API reference; remove fields this route does not support.
Authentication or balance issue
Check the Authorization bearer token and confirm the available balance in the console.
Context length exceeded
Prompt tokens exceed the Claude Sonnet 5.5 context window. Trim or retrieve only the relevant evidence and reuse cached prefixes.
Rate limited (429)
Back off and retry with jitter; batch or queue requests instead of sending parallel bursts.
Content or tool call rejected
Review sensitive content, malformed tool-call arguments, and JSON schema mismatches before retrying.
Claude Sonnet 5.5 API integration checks
Keep one EvoLink client and API key. Most requests carry over unchanged; the four steps below cover the settings that behave differently on Sonnet 5.5.
01
Switch the model ID
Point the client at EvoLink, keep the same API key, and select claude-sonnet-5-5. Both the OpenAI-compatible Chat Completions endpoint and the Anthropic Messages endpoint accept it.
Low migration
02
Replace thinking-off settings
To turn off up-front thinking, use thinking type between_tools at effort high or below. Progress notes between tool calls can still arrive as thinking blocks, so parse content by block type.
Breaking change
03
Re-run the effort sweep
The default is still high, but Anthropic recalibrated the levels, so a Sonnet 5 setting does not carry over one-to-one. Start at medium for agentic coding and multistep tool use and at low for chat, extraction, and classification.
Tune
04
Route tasks by completed-task value
Compare success rate, retries, output tokens, cache hits, wall-clock time, and human correction against Claude Sonnet 5 and Opus 5.5 on replayed production samples, not token price alone.
Evaluate
Prompt caching for repeated agent context
Cache stable system prompts and tool schemas to reuse context across agent turns. Writes and reads are billed separately. Anthropic lowered the minimum cacheable prompt length compared with Sonnet 5, so shorter prompts can now be cached. Use EvoLink’s displayed write and read prices to budget for cache hits and misses.
1M-token context with a 128K max output
Keep connected code, specifications, documents, and agent state in one working context. Thinking counts toward max_tokens, so leave room for it: for long agentic coding turns, 64K is a reasonable starting point.
thinking: disabled is rejected — between_tools is the lowest setting
Anthropic rejects thinking: disabled on Sonnet 5.5. EvoLink converts that legacy setting to adaptive thinking for compatibility. To turn off up-front thinking, use between_tools at effort high or below, with no other fields inside thinking. Progress notes between tool calls can still arrive as thinking blocks.
Forced tool choice is not supported
Forced tool choice is unsupported: Messages any/tool and Chat Completions required/named functions return an error. With auto, clearly request the tool and verify that a call occurred; auto does not guarantee one. Notes between tool calls may arrive as thinking blocks, so parse content by block type.
Claude Sonnet 5.5: five billing dimensions, metered separately
Claude Sonnet 5.5 does not bill as a single blended rate. Knowing which dimension a workload consumes, and that thinking is billed as output, keeps a high-volume budget predictable.
Input and output tokens
$1.900 per 1M input tokens and $9.500 per 1M output tokens on EvoLink, compared with Anthropic’s $2.000 and $10.000. Thinking is billed as output even when its text is not returned, so size max_tokens for the thinking plus the reply.
5-minute cache writes and cache reads
Storing a prefix for 5 minutes costs $2.375 per 1M tokens; reusing it costs $0.192 per 1M tokens. Anthropic’s official cache-read rate is 10% of its input rate. EvoLink rates are rounded, so use the displayed prices when estimating savings.
Web search tool
On routes that support web_search, each search costs $0.010. Search-result tokens are billed as ordinary input. Confirm route support before integrating; a configured price does not mean every route supports the tool.
Compare Claude routes after workload testing
EvoLink
First check whether Sonnet 5.5 reduces retries and review effort on your own tasks at the effort level you plan to run. Then compare price, context, caching, and workload fit to choose the production route.
Fixing failing tests and multi-file changes in a real repository, pull-request review, agents that call search and internal tools, and computer-use workflows — the everyday work where a Sonnet-rate route should finish the task without escalating to Opus.
The previous Sonnet at the same official list price; keep it where a workload is already validated and not yet re-tested.
Anthropic’s newest Opus, the step up for the hardest long-horizon coding, knowledge work, and visual analysis.
Claude Sonnet 5.5
Input / output$1.9 / $9.5
Context1M
CachingRead + write
Best forFixing failing tests and multi-file changes in a real repository, pull-request review, agents that call search and internal tools, and computer-use workflows — the everyday work where a Sonnet-rate route should finish the task without escalating to Opus.
Which model does this Claude Sonnet API page cover?
This page covers Claude Sonnet 5.5, available on EvoLink under model ID "claude-sonnet-5-5". Sonnet names a model family; select the exact version in your request rather than sending "sonnet" as the model ID. Earlier versions have separate model pages and rates.
What model ID should I use for the Claude Sonnet 5.5 API?
Send model "claude-sonnet-5-5". The same ID works for both the OpenAI-compatible Chat Completions endpoint and the Anthropic Messages endpoint.
How much does the Claude Sonnet 5.5 API cost on EvoLink?
EvoLink pricing: $1.900 per 1M input tokens and $9.500 per 1M output tokens, against Anthropic’s $2.000 and $10.000.
Is Claude Sonnet 5.5 more expensive than Claude Sonnet 5?
Anthropic lists both at the same per-token prices. EvoLink shows each model’s current rate on its own page, and the two can differ, so compare the displayed rates. Anthropic also reports that Sonnet 5.5 finishes agentic coding work in far fewer model requests than Sonnet 5; that is Anthropic’s measurement on its own tasks, so compare cost per accepted task on your workload at the effort you plan to run.
How is Claude Sonnet 5.5 prompt caching billed?
5-minute cache writes cost $2.375 per 1M tokens and cache reads cost $0.192 on EvoLink, compared with Anthropic’s $2.500 and $0.200. They are metered separately. Use the displayed rates, including rounding, to estimate your workload.
How is the Claude Sonnet 5.5 web search tool billed?
On routes that support web_search, each search costs $0.010. Search-result tokens are billed as ordinary input. Confirm route support before integrating; a configured price does not mean every route supports the tool.
Can I turn thinking off on Claude Sonnet 5.5?
Not with thinking: disabled: Anthropic rejects it, and EvoLink converts it to adaptive thinking. The lowest setting, between_tools, turns off up-front thinking; progress notes between tool calls can still arrive as thinking blocks. It requires effort high or below, no other fields inside thinking, and no effort change mid-conversation. Anthropic recommends trying adaptive thinking at low effort first.
Which effort level should I use on Claude Sonnet 5.5?
The default is high, but the levels are recalibrated from Sonnet 5, so set effort explicitly and re-test. Anthropic suggests starting at medium for agentic coding and multistep tool use and at low for chat, content generation, classification, and extraction. Reserve xhigh and max for work with a measured quality gain.
What changes when I migrate from Claude Sonnet 5?
Change the model ID, replace thinking: disabled with low effort or between_tools, and re-run the effort sweep. Forced tool choice is unsupported in both protocols (Messages any/tool; Chat Completions required/named functions). Thinking blocks from Sonnet 5.5 are read only by Sonnet 5.5, and on the Claude API computer use requires the computer toolset. Context, max output, and the tokenizer are unchanged.
Should I switch from Claude Sonnet 5 or Claude Opus 5.5 to Sonnet 5.5?
From Sonnet 5: pilot Sonnet 5.5 on agentic coding and tool-heavy agents, and keep Sonnet 5 as the rollback route until a matched replay at your chosen effort shows a lower cost per accepted task. From Opus 5.5: move everyday coding and high-volume agents where Sonnet 5.5 holds quality, and keep Opus 5.5 for the hardest long-horizon work, which Anthropic still recommends an Opus model for.
Which protocols and SDKs work with Claude Sonnet 5.5?
Use the OpenAI-compatible Chat Completions endpoint or the Anthropic Messages endpoint with the same EvoLink API key. Claude Code and other Anthropic-native clients work by pointing the base URL at EvoLink.
Do Claude Code subscription usage limits apply to the Claude Sonnet 5.5 API on EvoLink?
Claude subscription quotas are separate from EvoLink API billing. Requests through EvoLink consume your EvoLink balance at the displayed rates and remain subject to your account and route limits. Configure an EvoLink API key and the documented endpoint in a compatible client.
What context window and max output does Claude Sonnet 5.5 support?
The EvoLink route records a 1,000,000-token context window with up to 128K output tokens per response. Thinking counts toward max_tokens.
Why does my streaming UI go quiet between tool calls on Sonnet 5.5?
On Sonnet 5.5 longer notes the model writes between tool calls arrive as progress-update thinking blocks instead of text blocks, and their text is empty at the default display setting. With adaptive thinking, set thinking.display to “updates” (beta, header thinking-display-updates-2026-08-18) or “summarized”; with between_tools the notes come back with their text. Render each non-empty thinking block before the tool call it precedes, and pass the blocks back unchanged.
What should a production Claude Sonnet 5.5 evaluation measure?
Track first-pass success, accepted deliverables, retries, model requests per task, output tokens at your chosen effort, cache hit rate, valid tool calls, time to accepted result, human correction, and fallback rate — not the per-token rate alone.
