Effort: max · Max, Default Fallback
anthropicReleased Sep 22, 2026
Anthropic: Claude Opus 5.5
Always-on adaptive thinking defaults to medium effort; built for sustained work across large codebases and professional documents.
Source description
Claude Opus 5.5 is Anthropic's flagship model for demanding reasoning, coding, and long-horizon agentic work, succeeding Claude Opus 5. It is particularly strong at multi-step changes in large codebases, code...
Context window1M
Max output128K
Provider endpoints14
Benchmark results19
Agent runs1
Measured performance
All leaderboards ↗Benchmark results
Effort: max · Max, Default Fallback
GDPval-AA ↗2.1
1,866.27Effort: max · Max, Default Fallback
AutomationBench-AA ↗AA Index 4.3.2 collection
69.54%Effort: max · Max, Default Fallback
Effort: max · Max, Default Fallback · Harness unreported
SciCode ↗AA Index 4.3.2 collection
66.9%AA · Under reviewEffort: max · Max, Default Fallback
Humanity’s Last Exam ↗AA Index 4.3.2 collection
61.35%Effort: max · Max, Default Fallback
GDP.pdf ↗AA Index 4.3.2 collection
26.2%Effort: max · Max, Default Fallback
CritPt ↗AA Index 4.3.2 collection
31.71%AA · Under reviewEffort: max · Max, Default Fallback
AA-Omniscience ↗AA Index 4.3.2 collection
46.42Effort: max · Max, Default Fallback
AA-LCR ↗1.1
84.67%Effort: max · Max, Default Fallback
LiveBench Overall ↗2.0 · 2026-10-14 (dataset 2026-09-02)
86.53%Effort: max · single run · equal mean of 7 suites
LiveBench Design ↗2.0 · 2026-10-14 (dataset 2026-09-02)
86.9%Effort: max · single run
LiveBench Research ↗2.0 · 2026-10-14 (dataset 2026-09-02)
85%Effort: max · single run
LiveBench Browser Use ↗2.0 · 2026-10-14 (dataset 2026-09-02)
80%Effort: max · single run
LiveBench Coding ↗2.0 · 2026-10-14 (dataset 2026-09-02)
96.3%Effort: max · single run
LiveBench Agentic ↗2.0 · 2026-10-14 (dataset 2026-09-02)
85%Effort: max · single run
LiveBench Data Analysis ↗2.0 · 2026-10-14 (dataset 2026-09-02)
80.3%Effort: max · single run
LiveBench Reasoning ↗2.0 · 2026-10-14 (dataset 2026-09-02)
92.2%Effort: max · single run
Provider measurements
1 configurationSpeed
≈10K input tokens1 requestmedian / 72h
Artificial Analysis · first-party providers · serving endpoint not reported
Anthropic
Claude Opus 5.5 (Max, Default Fallback)
Reasoning max
- Output rate
- 97.4598 tok/s
- Time to first chunk
- 728.268 s
Last checked Source ↗
Methodology 2.2.0 & measurement details
First chunk includes request processing and network time, and may include reasoning.
- Prompt
- long · ≈10,000 input tokens
- Parallel requests
- 1
- Minimum answer
- 1,500 tokens
- Aggregation
- median over 72 hours
- Token counting
- o200k_base
- Benchmark client
- Google Cloud us-central1-a
- Serving endpoint
- Not reported
- Serving region
- Not reported
- Sample size
- Not reported
- Window end
- Not reported
Model + harness
All agent results ↗Agent runs
Terminal-Bench 4.0
claude-code2.1.284
Reasoning effort max
Tested model identity
Opus 5.5anthropic/claude-opus-5-5Pass rate64.85%
95% CI 61.74%–67.96%
n = 330Checked
Access & cost
14 quotesPrices across providers
Your token mix
USD · listed token ratesOpenRouter
Catalog quote- Input
- $4.00
- Output
- $20.00
- Cached input
- $0.20
- Input
- $4.00
- Output
- $20.00
Experiential
Standard list rate
- Input
- $4.00
- Output
- $20.00
- Cached input
- $0.20
Catalog list rate. Plan allowances and temporary discounts are separate.
- Input
- $4.00
- Output
- $20.00
- Cached input
- $0.20
Cache write $5/M tokens · 1h cache write $8/M tokens · Web search $0.01/search
- Input
- $4.00
- Output
- $20.00
- Cached input
- $0.20
Cache write $5/M tokens · 1h cache write $8/M tokens · Web search $0.01/search
- Input
- $4.00
- Output
- $20.00
- Cached input
- $0.20
Cache write $5/M tokens · 1h cache write $8/M tokens · Web search $0.01/search
- Input
- $4.00
- Output
- $20.00
- Cached input
- $0.20
Cache write $5/M tokens · 1h cache write $8/M tokens · Web search $0.01/search
- Input
- $4.00
- Output
- $20.00
- Cached input
- $0.20
Cache write $5/M tokens · 1h cache write $8/M tokens · Web search $0.01/search
- Input
- $4.40
- Output
- $22.00
- Cached input
- $0.22
Cache write $5.5/M tokens · 1h cache write $8.8/M tokens · Web search $0.01/search
- Input
- $4.40
- Output
- $22.00
- Cached input
- $0.22
Cache write $5.5/M tokens · 1h cache write $8.8/M tokens · Web search $0.01/search
- Input
- $4.40
- Output
- $22.00
- Cached input
- $0.22
Cache write $5.5/M tokens · 1h cache write $8.8/M tokens · Web search $0.01/search
- Input
- $4.40
- Output
- $22.00
- Cached input
- $0.22
Cache write $5.5/M tokens · 1h cache write $8.8/M tokens · Web search $0.01/search
- Input
- $4.40
- Output
- $22.00
- Cached input
- $0.22
Cache write $5.5/M tokens · 1h cache write $8.8/M tokens · Web search $0.01/search
- Input
- $8.00
- Output
- $40.00
- Cached input
- $0.40
Cache write $10/M tokens · 1h cache write $16/M tokens · Web search $0.01/search
Model specifications
- Developer
- anthropic
- Modality
- Language
- Context
- 1,000,000 tokens
- Maximum output
- 128,000 tokens
- Weights
- Not reported
- Released
- Sep 22, 2026
- Provider listed
- Sep 22, 2026
- First discovered
- Oct 16, 2026
Exact identities
- openrouter
anthropic/claude-opus-5.5 - deepinfra
anthropic/claude-opus-5-5 - experiential
claude-opus-5.5 - artificial-analysis
2f3c4dc9-a450-4303-8697-0237257cf08f - livebench
claude-opus-5.5
Sources
Updated Oct 18, 2026
More from anthropic
View all ↗anthropicListed 10 days ago
Anthropic: Claude Haiku 5.5
Adds adjustable effort to the Haiku tier for compaction, browser tasks and narrowly scoped subagents.
1M contextImage inputTool callingNo reported score
No reported score
$0.10in$0.50out/ 1M tokens · USDAmazon Bedrock · via OpenRouter · 12 endpointsamazon-bedrock · prompt ≤ 100000 tokens
anthropicListed 10 days ago
Anthropic: Claude Haiku 5.5 (batch)
Claude Haiku 5.5 is Anthropic's small, fast model for high-volume, cost-sensitive work such as summarization, subagents, and browser use. It succeeds Claude Haiku 4.5 with stronger coding, computer use, and...
1M contextImage inputTool callingNo reported score
No reported score
$0.05in$0.25out/ 1M tokens · USDAnthropic · via OpenRouter · 2 endpointsanthropic · prompt ≤ 100000 tokens
anthropicListed 19 days ago
Anthropic: Claude Sonnet 5.5 (batch)
Claude Sonnet 5.5 is Anthropic's Sonnet-class model for well-scoped everyday work, succeeding Claude Sonnet 5 as a direct upgrade. It is especially strong at building features, fixing bugs, and producing...
1M contextImage inputTool callingNo reported score
No reported score
anthropicReleased 20 days ago
Anthropic: Claude Sonnet 5.5
Tuned for quick iteration on scoped fixes and documents; Anthropic favors Opus 5.5 for open-ended work needing sustained judgment.
1M contextImage inputTool calling