# Claude Sonnet 5.5 Claude Sonnet 5.5 is an AI model released by Anthropic on Sep 28 2026. It has a 1M context window. Tracked results include 70.6% on Terminal-Bench 4.0, 64.5% on Humanity's Last Exam (with tools) and 55.5% on CursorBench 4.0. ## Facts | Field | Value | | --- | --- | | Model | Claude Sonnet 5.5 | | Developer | Anthropic | | Release date | Monday, Sep 28 2026 | | Licensing | Closed | | Context window | 1M | ## API pricing All rates in USD per 1,000,000 tokens, pay-as-you-go. | Tier | Input | Cached input | 5 min cache write | 1 hr cache write | Output | | --- | --- | --- | --- | --- | --- | | All documented contexts | $2.00 | $0.20 | $2.50 | $4.00 | $10.00 | Same rates as Sonnet 5; Anthropic said it needs fewer tokens per task, for up to 30% lower cost per task in its testing. Verified September 28, 2026 against the first-party source: https://platform.claude.com/docs/en/about-claude/pricing ## Tracked benchmark scores | Benchmark | Score | Source | What it measures | | --- | --- | --- | --- | | CursorBench 4.0 | 55.5% | [CursorBench](https://cursor.com/cursorbench), retrieved 2026-09-28 | Cursor's own test of coding agents on ambiguous, multi-file tasks taken from real Cursor sessions — editing, refactoring, investigating a codebase, understanding what the user meant, managing jobs and following a design. Cursor runs each model at several reasoning efforts; each release here carries the score of its best listed effort. Scores aren't comparable with earlier CursorBench versions. Higher is better. | | FrontierCode v1.1 (Main) (main split) | 52.1% | Lab | A set of very hard, frontier-difficulty coding tasks an AI agent has to complete end to end. The score is the share of tasks in the main split it solves. Higher is better. | | Terminal-Bench 4.0 | 70.6% | Lab | Can the AI work in a command-line terminal — running commands and finishing technical setup tasks the way a developer would? Version 4.0 recalibrated how much time, CPU and memory each task gets, removed eight tasks and fixed nineteen, so fewer runs fail for reasons that have nothing to do with the model. Scores are not comparable with earlier versions. Higher is better. | | Humanity's Last Exam (with tools) | 64.5% | Lab | Humanity's Last Exam — extremely hard expert questions across many subjects. “With tools” means the AI is allowed to search the web or run code while answering. Higher is better. | | GDPval-AA v2.1 | 1844 | Lab | economically valuable knowledge work (v2.1, Crowd-BT Elo fit) | | AA-Briefcase v1.1 | 1811 | Lab | Artificial Analysis agentic office-work eval (Elo, v1.1 rating fit) | | Chartography (no tools) | 61.6% | Lab | The same chart-centred test with no tools: the AI has to read each chart unaided. Scores run far lower than the with-tools version, so read the two as separate tests. Higher is better. | ## About Claude Sonnet 5.5 Claude Sonnet 5.5, released September 28, 2026, was the second model of Anthropic's 5.5 generation, six days after Claude Opus 5.5 and three months after Claude Sonnet 5 opened the 5-series at the mid tier. Anthropic positioned it as the faster, cheaper complement to Opus 5.5: strongest at well-scoped everyday work such as bug fixes and polished documents, slides and spreadsheets, while Opus 5.5 kept the complex, open-ended tasks that need sustained judgment. The price stayed at Sonnet 5's $2 per million input tokens and $10 per million output, with cache reads at $0.20, but Anthropic said the model needed far fewer tokens for the same work, for up to 30% lower cost per task, and generated output more than 30% faster, which made it the fastest Sonnet at the time. It shipped with a 1-million-token context window, 128,000 tokens of maximum output and a June 2026 knowledge cutoff, and was the first Sonnet to launch with cyber safeguards, because its cybersecurity capabilities were comparable to Opus 5's. The launch table's biggest jump was on agentic coding: 70.6% on Terminal-Bench 4.0 against Sonnet 5's 10.3%, and above the 66.4% Anthropic reported for Opus 5.5 at xhigh effort. On FrontierCode v1.1 it scored 52.1% at xhigh effort and 46.2% at max, against Sonnet 5's 42.4% and Opus 5.5's 54.4%; Anthropic attributed the drop at max effort to the model running more subagent code reviews that made edits outside the task's scope. On knowledge work it came within a few points of Opus 5.5, with a GDPval-AA v2.1 Elo of 1844 against 1846 and an AA-Briefcase v1.1 Elo of 1811 against 1822, both about 400 points above Sonnet 5. It also scored 64.5% on Humanity's Last Exam with tools, 80.1% partial completion on OSWorld 2.1 computer use and 61.6% on Chartography without tools, up from Sonnet 5's 15.6%. It was the first Sonnet model to beat Pokémon Red working only from screenshots. At launch it was available through Anthropic's API as claude-sonnet-5-5 and on Amazon Bedrock, Google Cloud Vertex AI and Microsoft Foundry, with Claude Haiku 5.5 announced to follow in the coming weeks. ## Questions and answers ### When was Claude Sonnet 5.5 released? Claude Sonnet 5.5 was released by Anthropic on Monday, Sep 28 2026. ### Who made Claude Sonnet 5.5? Claude Sonnet 5.5 was built by Anthropic. AI safety company building the Claude family of models. Founded in 2021 by former OpenAI researchers. ### How much does Claude Sonnet 5.5 cost? Claude Sonnet 5.5 costs $2.00 per million input tokens and $10.00 per million output tokens through the Anthropic API. Cached input is $0.20 per million tokens. Rates are pay-as-you-go API prices verified against Anthropic's published pricing on September 28, 2026. ### What benchmark scores did Claude Sonnet 5.5 get? Claude Sonnet 5.5 reports 7 tracked benchmark scores — CursorBench 4.0: 55.5%; FrontierCode v1.1 (Main) (main split): 52.1%; Terminal-Bench 4.0: 70.6%; Humanity's Last Exam (with tools): 64.5%; GDPval-AA v2.1: 1844; AA-Briefcase v1.1: 1811; Chartography (no tools): 61.6%. Tracked scores may come from lab reports or independent benchmarks; source details accompany the benchmark data. It holds the best score among all models tracked here on Terminal-Bench 4.0. ### What is the context window of Claude Sonnet 5.5? Claude Sonnet 5.5 has a context window of 1M. That is the maximum amount of input plus output the model can hold in a single request. ### Is Claude Sonnet 5.5 open source? No. Claude Sonnet 5.5 is a closed model. The weights are not published — it is available only through the provider's own API, apps, or partner platforms. ### What came before and after Claude Sonnet 5.5? Anthropic's previous tracked release was Claude Opus 5.5 on Sep 22 2026, 6 days earlier. It is the most recent Anthropic model tracked on AI Release Tracker. --- Canonical page: https://aireleasetracker.com/model/anthropic/claude-sonnet-5.5 Site index: https://aireleasetracker.com/llms.txt Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.