Learning Objectives
- Understand where Claude Sonnet 5.5 sits in the Claude model family
- Explain what changed from Sonnet 5, and what did not
- Decide when Sonnet 5.5 is enough and when to step up to Opus 5.5
What Is Claude Sonnet 5.5?
Claude Sonnet 5.5 is Anthropic's mid-tier Claude model, released on September 28, 2026 as the second model in the Claude 5.5 family, six days after Claude Opus 5.5. In the lineup it sits between the small, fast Claude Haiku 5.5, which completed the family on October 7, and the flagship tiers, Claude Opus and the premium Claude Fable.
On October 7, 2026 Anthropic also halved Sonnet 5.5's cache-read price, from 20 cents to 10 cents per million tokens. Input and output prices did not change, but cache reads make up most of the tokens on agentic and coding work, so Anthropic says the cut makes Sonnet 5.5 about 20 percent cheaper on most of those tasks.
Anthropic's own framing is a division of labour. Opus 5.5 is for complex work that needs careful judgment; Sonnet 5.5 is strongest at well-scoped everyday tasks, fixing bugs, and producing polished documents, slides and spreadsheets. It replaces Claude Sonnet 5, released June 30, 2026, which remains available to developers on the API.
π‘Key Concept
"Agentic" means the model can act, not just answer. An agentic model plans a multi-step task, uses tools like browsers and terminals, checks its own work, and adjusts, rather than producing a single response and stopping. Sonnet 5 introduced that emphasis to the Sonnet line; Sonnet 5.5 keeps it and makes it faster and cheaper per task.
What Changed From Sonnet 5
- Speed. Anthropic says Sonnet 5.5 generates output more than 30 percent faster than Sonnet 5, making it the fastest Sonnet model to date.
- Cost per task, not price per token. The list price did not move. The saving comes from the model needing far fewer tokens to finish the same work, which Anthropic puts at up to 30 percent less per task.
- Agentic coding and computer use. In Anthropic's own tables Sonnet 5.5 scores 70.6 percent on Terminal-Bench 4.0, slightly ahead of Opus 5.5's 66.4 percent, and 80.1 percent on OSWorld 2.1 against Opus 5.5's 81.8 percent.
- Knowledge work. Anthropic reports it lands two points below Opus 5.5 on GDPval-AA, a test of real-world work across many occupations.
- Writing. Like Opus 5.5, it writes more clearly than the previous generation, and early testers called it a better collaborator than Sonnet 5.
These are vendor-run benchmarks. Anthropic itself cautions that they capture only one facet of capability, and that in its own testing and that of outside testers, Opus 5.5 remains clearly stronger at open-ended work requiring sustained judgment.
Cyber Safeguards Now Apply to Sonnet
Sonnet 5 shipped with substantially reduced cybersecurity capabilities, a deliberate safety choice. Sonnet 5.5 reverses that. Anthropic says its cyber capabilities are comparable to Claude Opus 5, so it is the first Sonnet model to launch with the cyber safeguards and fallbacks Anthropic built for its most capable models. Its biology safeguards are the same as Sonnet 5's.
Both safeguards target a narrow set of high-risk requests, and Anthropic says routine software development and most life-sciences work are unaffected. In practice, security teams doing legitimate offensive testing may now see refusals or fallbacks on Sonnet that they did not see on Sonnet 5.
Pricing
- Sonnet models included (with usage limits)
- Projects and Artifacts
- Sonnet 5.5 included
- Higher usage limits
- Priority access
- 10 cents per million cached-read tokens (halved October 7, 2026)
- model id claude-sonnet-5-5
- Still available to developers
- model id claude-sonnet-5
Sonnet 5.5 costs the same per token as Sonnet 5, which never moved to the $3 input and $15 output rate once expected after August 31. Developers reach it with the claude-sonnet-5-5 model id, and it is available on Amazon Web Services, Google Cloud and Microsoft Azure as well as Anthropic's own platform. One migration detail: developers who run Sonnet with thinking switched off need to move to a new setting Anthropic calls between_tools before switching models.
Related Tools
- Claude Opus 5.5 β the stronger model for complex, open-ended work
- Claude Fable β Anthropic's premium tier for the hardest reasoning
- Claude Haiku 5.5 β the faster, cheaper tier below Sonnet; on Anthropic's Terminal-Bench 4.0 it scores 39.2 percent against Sonnet 5.5's 70.6 percent, so it suits narrow, high-volume tasks rather than hard agentic coding
- Claude.ai β the consumer interface where Sonnet is available on every plan
Strengths
- Faster and cheaper per task than Sonnet 5 at an unchanged per-token price
- Near-Opus results on agentic coding and computer use in Anthropic's own benchmarks, especially at lower effort settings where it costs much less per task
- Clear writing β polished documents, slides and spreadsheets are a stated strength
- Broad availability β every Claude plan, the API, and the three major clouds
Limitations and Considerations
- Not the top of the lineup β Anthropic says Opus 5.5 remains clearly stronger at complex, open-ended work requiring sustained judgment, and the premium Fable line sits above both
- Cyber safeguards now apply β legitimate security work may hit refusals or fallbacks that Sonnet 5 did not trigger
- Benchmarks are self-reported β the Terminal-Bench and OSWorld figures come from Anthropic's own tables, not independent testing
- A migration step for some developers β anyone running Sonnet with thinking off must adopt the new
between_toolssetting first
Key Takeaways
- Claude Sonnet 5.5 (released September 28, 2026) is Anthropic's mid-tier model and the second member of the Claude 5.5 family, which Haiku 5.5 completed on October 7; its cache-read price fell by half that day, to 10 cents per million tokens
- It runs more than 30 percent faster than Sonnet 5 at the same $2 input and $10 output per million tokens, and Anthropic says it costs up to 30 percent less per task
- It is the first Sonnet model to carry Anthropic's cyber safeguards, because its cybersecurity capability is comparable to Opus 5
- Opus 5.5 remains the better choice for complex, open-ended work; Sonnet 5.5 is built for well-scoped everyday tasks and fast iteration










