Claude Sonnet 5.5とは？料金と性能、上位モデルとの違いを比較

"Sonnet 5.5 is a faster, lower-cost complement to Claude Opus 5.5. Where Opus 5.5 is built for complex work requiring careful judgment, Sonnet 5.5 is strongest at well-scoped everyday tasks, fixing bugs, and creating polished documents, slides, and spreadsheets." ／ "benchmark scores capture only one facet of a model’s capabilities; in our own testing, and in that of external testers, Opus 5.5 remains clearly stronger at complex, open-ended work requiring sustained judgment." — Anthropic 公式発表の冒頭とベンチマークの節より

"Context window 1M tokens" ／ "Max output 128K tokens" ／ "Released September 28, 2026" ／ "Platforms Claude API Amazon Bedrock Google Cloud Microsoft Foundry Claude Platform on AWS" — Claude Docs のモデル概要ページより

"Sonnet 5.5 scores 70.6% on Terminal-Bench 4.0, an agentic coding evaluation, compared to Sonnet 5’s 10.3%. It scores two points below Opus 5.5 on GDPval-AA" ／ "It’s close to Opus 5.5 in computer use and chart recognition" ／ "On several benchmarks, Sonnet 5.5 at Low or Medium effort beats Sonnet 5’s best score for about a tenth of the cost per task." — Anthropic 公式発表の冒頭とベンチマークの節より（Opus 5.5 の Terminal-Bench 4.0 は発表の脚注で Xhigh の値＝同モデルの最高スコア）

"They found it to be a more natural conversational partner and remarked on its knack for design, noting that it adds polish to user interfaces and can follow slide templates to create decks that require minimal editing." ／ "Its speed also makes it well suited to fast iteration on less complex tasks." ／ "In one internal test, we gave it a public company’s quarterly earnings materials and call transcripts, along with a slide template, and asked for a 10-slide operating review. Two experts judged its first draft to be ready to send as is." ／ "It complements Opus 5.5 best when running at lower effort settings, where it costs less per task." ／ "When Claude Opus 5.5 sets the architecture and general framework for a game, I would feel confident in letting Sonnet 5.5 implement it." — Anthropic 公式発表の本文と導入企業の声（Creator・Kevin Ngo氏）より

"For agentic coding and multistep tool use, start at medium for well-specified tasks and move to high for harder or longer ones. For chat and other latency-sensitive work, start at medium or low." — Claude Docs の「Behavior differences」節より

"Claude Opus 5.5 1M 128K $4 / $20" ／ "Claude Sonnet 5.5 This model 1M 128K $2 / $10" — Claude Docs のモデル概要ページの比較表より

"Sonnet 5.5 is priced the same as Sonnet 5 at $2 per million input tokens, $10 per million output tokens, and $0.20 per million tokens for cache reads, but it typically needs far fewer tokens to do the same work. In our testing, it costs up to 30% less per task than its predecessor." ／ "Sonnet 5.5 generates outputs 30%+ faster than Sonnet 5, making it our fastest Sonnet model to date." — Anthropic 公式発表の料金と速度の節より

"5m cache write $2.50 / MTok" ／ "1h cache write $4 / MTok" ／ "Cache read $0.20 / MTok" ／ "Batch API 50% discount on input and output" — Claude Docs のモデル概要ページの Pricing 表より

"Claude Sonnet 5.5 has the same prices as Claude Sonnet 5, including prompt caching and batch processing rates." — Claude Docs の「Pricing」節より

"In Claude Code and our apps, the default effort is set to Medium, while the Claude Platform defaults to High. At lower settings, Claude answers faster and uses fewer tokens, which suits routine work. At higher settings, Claude reasons for longer and checks its work more thoroughly." — Anthropic 公式発表の effort の節より

"An effort level doesn't produce the same amount of thinking as it did on Claude Sonnet 5. Re-run your effort sweep rather than carrying a setting over." — Claude Docs の「Behavior differences」節より

"Five breaking changes affect code already running on Claude Sonnet 5:" ／ "Forced tool use returns an error" ／ "Thinking blocks are tied to the model and the conversation" ／ "The advisor tool rejects Claude Opus 4.8, Claude Opus 4.7, and Claude Sonnet 5 as advisors" ／ "It enforces that check by default for accounts created on or after August 31, 2026, 00:00 UTC" ／ "One more change alters the response shape without failing any request: text between tool calls comes back in thinking blocks." ／ "No other model reads Claude Sonnet 5.5 thinking blocks." — Claude Docs の冒頭と「Breaking changes」節より

"To turn off up-front thinking on Claude Sonnet 5.5, send thinking: {"type": "between_tools"} instead of "disabled"." ／ "between_tools is accepted at low, medium, and high effort. At xhigh or max effort, a request with between_tools returns a 400 error." — Claude Docs の「Breaking changes」節より

"tool_choice: type "tool" and "any" are not supported for this model." ／ "To make the model call a tool rather than reply in text, say in the prompt when the tool applies." ／ "On Amazon Bedrock, Claude Sonnet 5.5 accepts the earlier computer_20251124 tool." — Claude Docs の「Breaking changes」節より