What breaks when I switch from Sonnet 5?

Per Anthropic: thinking type "disabled" now returns an error (send "between_tools" instead, at high effort or below); forced tool use (tool_choice "any" or "tool") is not supported, so use "auto" with strict tool use; computer use on the Claude API needs computer_toolset_20260801; replaying thinking blocks after editing history can return a 400; and the advisor tool no longer accepts Opus 4.8, Opus 4.7, Sonnet 5 or Haiku 5.5 as advisors. Non-default temperature, top_p and top_k are also rejected by Anthropic on this model; RouteMux drops those three parameters before forwarding, so SDKs that set them by default keep working.

Should I use Sonnet 5.5 or Sonnet 5?

Start new work on Sonnet 5.5. Sonnet 5 remains available if an existing integration relies on disabling thinking, forced tool calls or computer_20251124 and you cannot change that yet; Anthropic commits to retiring it no sooner than June 30, 2027.

How do Sonnet 5.5 and Opus 5.5 divide the work?

In Anthropic's framing, Sonnet 5.5 suits well-scoped everyday tasks, bug fixes and polished documents, while Opus 5.5 is built for complex work that needs careful judgment. Anthropic reports that Sonnet 5.5 at max effort is comparable to Opus 5.5 on several evaluations, but Opus 5.5 remains clearly stronger on open-ended work that needs sustained judgment.

Claude Sonnet 5.5

Save 72%Official

anthropic/claude-sonnet-5-5

Anthropic's September 2026 Sonnet — successor to Sonnet 5, which Anthropic calls the best balance of speed and intelligence.

Get an API key View code sample

Live model playground

Try Claude Sonnet 5.5 here

Tune the request, see the current price ceiling, and inspect the real response without leaving this model page.

streamchat

1Configure input

Prompt

177 / 20,000

Add images

Maximum estimated cost—

2Live resultExample

Claude Sonnet 5.5

Work in rates (tank per hour):

Inlet fills at 1/6 per hour.
Outlet drains at 1/9 per hour.
Net rate with both open: 1/6 − 1/9 = 3/18 − 2/18 = 1/18 per hour.

Time to fill = 1 ÷ (1/18) = 18 hours.

Context window: 1M
Max output: 128K
Input modalities: TextImage
Output modalities: Text
Capabilities: Vision · Reasoning · Tool calling · Coding · Prompt cache · Paid cache writes
Upstream release: 2026-09-28
RouteMux name: anthropic/claude-sonnet-5-5

What is Claude Sonnet 5.5

Claude Sonnet 5.5 was released by Anthropic on September 28, 2026, succeeding Sonnet 5, as the second model in the Claude 5.5 family, with retirement not sooner than September 28, 2027. Anthropic describes it as the best combination of speed and intelligence, and the models overview rates its comparative latency as Fast. Anthropic says it is strongest at well-scoped everyday tasks, fixing bugs and creating polished documents, slides and spreadsheets, while Opus 5.5 remains clearly stronger at complex, open-ended work that needs sustained judgment. Its output is more than 30% faster than Sonnet 5.

It has a 1M-token context window, up to 128K output tokens per request, image input and tool calling, a reliable knowledge cutoff of June 2026, and the same tokenizer as Sonnet 5. Adaptive thinking is on by default, the API default effort is high, and all five levels are recalibrated. Anthropic's starting points: high for general workloads; medium for well-specified agentic coding and multistep tool use, moving to high for harder or longer tasks; medium or low for chat and latency-sensitive work.

Moving over from Sonnet 5 is mostly a model-name change, but Anthropic lists five request shapes that now return an error; they are covered in the FAQ below. Separately, longer notes the model writes between tool calls now arrive in thinking blocks that are empty by default; to show them, set display or switch to between_tools.

Through RouteMux, Sonnet 5.5 costs $0.56 per 1M input tokens against Anthropic's $2, and $2.8 per 1M output tokens against $10.

Before you choose

What breaks when I switch from Sonnet 5?: Per Anthropic: thinking type "disabled" now returns an error (send "between_tools" instead, at high effort or below); forced tool use (tool_choice "any" or "tool") is not supported, so use "auto" with strict tool use; computer use on the Claude API needs computer_toolset_20260801; replaying thinking blocks after editing history can return a 400; and the advisor tool no longer accepts Opus 4.8, Opus 4.7, Sonnet 5 or Haiku 5.5 as advisors. Non-default temperature, top_p and top_k are also rejected by Anthropic on this model; RouteMux drops those three parameters before forwarding, so SDKs that set them by default keep working.
Should I use Sonnet 5.5 or Sonnet 5?: Start new work on Sonnet 5.5. Sonnet 5 remains available if an existing integration relies on disabling thinking, forced tool calls or computer_20251124 and you cannot change that yet; Anthropic commits to retiring it no sooner than June 30, 2027.
How do Sonnet 5.5 and Opus 5.5 divide the work?: In Anthropic's framing, Sonnet 5.5 suits well-scoped everyday tasks, bug fixes and polished documents, while Opus 5.5 is built for complex work that needs careful judgment. Anthropic reports that Sonnet 5.5 at max effort is comparable to Opus 5.5 on several evaluations, but Opus 5.5 remains clearly stronger on open-ended work that needs sustained judgment.

Claude Sonnet 5.5 API pricing

Prices verified 2026-10-06

InputOfficial $2 / M tokens

$0.56/ M tokensSave 72%

OutputOfficial $10 / M tokens

$2.8/ M tokensSave 72%

Cache read $0.056/M · Cache write $0.7/M

Estimate cost

Input tokensOutput tokens

Estimated cost$0.0084

Key capabilities

between_tools is the lowest thinking setting: To turn off up-front thinking, send thinking "between_tools", valid at low, medium and high effort; "disabled" returns a 400. Progress text between tool calls still comes back.
Recalibrated effort: A given level does not produce the same amount of thinking as on Sonnet 5, so Anthropic recommends a fresh sweep on your own evals rather than carrying settings over.
Change instructions mid-conversation: Supports system messages inside messages, mid-conversation tool changes (beta) and per-message effort (beta), none of which Sonnet 5 has. The minimum cacheable prompt drops to 512 tokens.
Thinking blocks bound to the conversation: Replaying 5.5 thinking blocks after editing earlier system, tools or messages can return a 400. Keep conversations append-only and change instructions with mid-conversation system messages.

Limited time10% extra Credits on your first top-up — up to 200 bonus Credits

Applies to your first top-up within 72 hours of signing up. Minimum $20, once per account.

Interface and protocols

Anthropic MessagesStreaming

Use any of these IDs to call this model via the API.

RouteMux nameanthropic/claude-sonnet-5-5

Replace the ROUTEMUX_KEY placeholder with your API key. Create one →

from anthropic import Anthropic

client = Anthropic(
    base_url="https://api.routemux.com/anthropic",
    api_key="<ROUTEMUX_KEY>",
)

message = client.messages.create(
    model="anthropic/claude-sonnet-5-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "What is the meaning of life?"}],
)
print(message.content[0].text)

base_url: https://api.routemux.com/anthropic

Performance (last 30 days)

9,351 requests · last 30 days

Total usage1156.80M

P50 first-token latency3.76s

Avg throughput47 tok/s

Success rate100.0%

Uptime · 24htested via anthropic_messages

100.00%

Success rate · 60d · from real requests

Claude Sonnet 5.5 FAQ

›What is Claude Sonnet 5.5?

Anthropic's September 2026 Sonnet — successor to Sonnet 5, which Anthropic calls the best balance of speed and intelligence.

›How large is the context window of Claude Sonnet 5.5?

Claude Sonnet 5.5 supports a context window of up to 1M tokens, with up to 128K output tokens per request.

›How much does the Claude Sonnet 5.5 API cost?

Through RouteMux, Claude Sonnet 5.5 costs $0.56 for input and $2.8 for output (USD per 1M tokens), below the official list price.

›How do I call Claude Sonnet 5.5 via API?

Claude Sonnet 5.5 is OpenAI-compatible: point your base URL at RouteMux and set the model field to anthropic/claude-sonnet-5-5 — no code changes needed.

›Which input and output modalities does Claude Sonnet 5.5 support?

Claude Sonnet 5.5 accepts text, image as input and produces text as output.

›When was Claude Sonnet 5.5 released?

Claude Sonnet 5.5 was released on September 28, 2026.

Sources

Reviewed by the RouteMux team on 2026-10-24

View as Markdown