Claude Opus 5.5
Claude Opus 5.5Rankingat max effort?
Full ranking →Alternatives?
CheaperClaude Sonnet 5.5max+12%-9%$5.46No forced tool use+5
OpenGLM 5.3 Flashmax-23%-95%$0.28Defaults to max · Self-hosted+4CheaperClaude Haiku 5.5max-5%-96%$0.21Remove sampling parameters+5CheaperClaude Opus 5high-11%-40%$3.61BetterClaude Fable 5.1max+7%+28%$7.63Replace forced tool_choice+5Claude Sonnet 5.5max+12%-9%$5.46No forced tool use+5Claude Haiku 5.5max-5%-96%$0.21Remove sampling parameters+5Claude Opus 5high-11%-40%$3.61Claude Sonnet 5.5max+12%-9%$5.46No forced tool use+5Claude Fable 5.1max+7%+28%$7.63Replace forced tool_choice+5GLM 5.3 FlashmaxOpen-23%-95%$0.28Defaults to max · Self-hosted+4GLM 5.3maxOpen-28%-66%$2.01No image input+5
Kimi K3maxOpen-43%-67%$2.00Sampling parameters are fixed+3
Qwen3.8 27BxhighOpen-67%-83%$1.01Context: 1M → 262K+5
Muse Glimmer 30BhighOpen-70%-99%$0.058Context: 1M → 131KClaude Sonnet 5.5max+12%-9%$5.46No forced tool use+5Claude Fable 5.1max+7%+28%$7.63Replace forced tool_choice+5Claude Fable 5max-3%+46%$8.75Claude Haiku 5.5max-5%-96%$0.21Remove sampling parameters+5Claude Opus 5max-17%same$5.86Claude Sonnet 5max-18%-15%$5.09Claude Opus 4.8max-32%-32%$4.08Claude Opus 4.7high-57%—$5 / $25All Anthropic models →History
9 versionsBenchmarks
in Chat & KnowledgeWatch out 6
No forced tool use›
Forced tool_choice any or tool returns HTTP 400; remove the forced choice. Source ↗
Thinking is always on›
Disabled thinking or a manual budget returns HTTP 400; omit thinking or use adaptive thinking. Source ↗
Remove sampling parameters›
Non-default temperature, top_p or top_k returns HTTP 400; omit these parameters and use prompting. Source ↗
No assistant prefill›
Assistant prefill returns HTTP 400; use structured outputs or system instructions, with system instructions on Bedrock. Source ↗
Old computer tool rejected · Claude API / Google Cloud›
The computer_20251124 tool returns HTTP 400 on the Claude API and Google Cloud; use computer_toolset_20260801 there. Source ↗
Return thinking blocks unmodified›
Edited, reordered or partially dropped thinking blocks return HTTP 400; echo assistant messages complete and unmodified with tool results. Source ↗
FAQ
How much does Claude Opus 5.5 cost?›
$4 per 1M input tokens and $20 per 1M output tokens; cached input $0.20 per 1M. A typical benchmark task costs about $5.98 at max effort.
Is Claude Opus 5.5 free to use?›
There is no free API; the cheapest paid price is $4 per 1M input tokens.
How good is Claude Opus 5.5, and what is it best at?›
#1 of 334 in Chat & Knowledge, #2 of 132 in Coding, #5 of 133 in Agents & Tool Use, #1 of 195 in Writing & Language, #1 of 124 in Vision & Documents. Strongest at: Coding, Agents & work tasks, Reasoning, Long context, Knowledge, Vision.
Is Claude Opus 5.5 better than Claude Opus 5?›
On MLCR-AA, Claude Opus 5.5 scores 66.7 vs 59.4 — higher. It is 66% more expensive per task.
How can I use Claude Opus 5.5?›
Through the API of Amazon Bedrock, Anthropic, Azure, Claude Platform on AWS, DeepInfra and 1 more.
What is the context window of Claude Opus 5.5?›
1M tokens, with up to 128K output tokens.
When was Claude Opus 5.5 released? Is there a newer version?›
Released 2026-09-22. It is the current version in its line.
Is Claude Opus 5.5 open source?›
No, Claude Opus 5.5 is a closed model available through an API. The closest open-weights option here is GLM 5.3 Flash.
Can Claude Opus 5.5 read images?›
It accepts text and images, PDF files as input.
What is Claude Opus 5.5?›
Claude Opus 5.5 is a closed language model from Anthropic, released Sep 2026, the successor to Claude Opus 5.
What are the best alternatives to Claude Opus 5.5?›
Claude Sonnet 5.5 (cheaper), GLM 5.3 Flash (open weights).