Claude Opus 5.5
anthropic/claude-opus-5.5- Anthropic
- Family · Claude Opus 5
- Released 22 Sep 2026
- Text · Image · File → Text
- Reasoning · adjustable
- Tools · JSON · Vision
Spec & list price
Context1M
Max output128K
Input / 1M$4.00
Output / 1M$20.005× input
Scores · standing
Intelligence58.4#2/489
LiveBench85.1#3/32
Usage
Tokens / day655.3B▼ 21% 7d
Usage rank#8
Requests / day3.7M
Tool calls / req0.66
Per request · 30 days
Avg input123.0K8.3× median
Avg output1.2K1.7× median
Cache hit90%1.3× median
Reasoning share51%about median
Tool-call failures0.5%0.3× median
Consumption
Peak 1.5T 23 OctLow 187.7B 26 SepRank today #8
News & trends
567 mentions this week ▲ 688%this model · videos, posts, articles · 30 days
Trending · 14 days
- GPT-6.1 Terra vs. Claude Opus 5.5: Cheaper and faster in three tests at max1 points
- Claude Opus 5.5 Brought NBA 2K27 to Roblox...68.4K views
- @ArtificialAnlys on X128.3K views
- @simonw on X119.5K views
- I Tested China's New Model vs Claude Opus 5.5 (Kimi K3.1)77.2K views
- Claude Opus 5.5 helps developer bring the entire Super Mario 64 into mixed reality, running natively in full 3D at 90 FPS, with the game world appearing right inside his room.3.6K upvotes
- Claude Opus 5.5 is insane in creating creating 3D movies.2.7K upvotes
News
- GPT-6.1 Terra vs. Claude Opus 5.5: Cheaper and faster in three tests at maxthenewstack.io · 2 Nov
- llama.cpp b12027llama.cpp · 1 Nov
- llama.cpp b12019llama.cpp · 31 Oct
- llama.cpp b11976llama.cpp · 29 Oct
Benchmarks
bar = share of the top score · Benchmark data: Artificial AnalysisSpeed · median across 10 providers, latest day
Output speed#125/33481 tok/s1.2× faster than median
Fastest provider222 tok/sbest of 10 providers
First token#2923.02 s2.9× longer than median
First answer token#101455 sthinking included
#2 cheapest of 3 with a similar score
Value
Intelligence Index
Blended price per 1M tokens (log)
Family progress
Intelligence Index
- Family releases58.4
LiveBench suites
Browser88.4
Coding96.3
Data Analysis80.3
Design86.9
Long Agentic72.8
Reasoning92.2
Research85.0
▲ Strongest: Coding, Reasoning. ▼ Weakest: Long Agentic (73).
Technical details
- Reasoning
- on · high
- Tokenizer
- Claude
- Moderation
- moderated
Reasoning effort
- max
- xhigh
- high · default
- medium
- low
Request parameters 11
- tools
- tool_choice
- response_format
- structured_outputs
- reasoning
- reasoning_effort
- include_reasoning
- max_tokens
- max_completion_tokens
- stop
+1 more
- verbosity
When to use
Best for
- Long autonomous agent runs
- Coding agents on large codebases
- Expert science and exam-style reasoning
- Design and front-end generation
- Research agents over large context
Think twice for
- High-volume simple chat (cheaper options exist)
- Browser automation (its weakest suite)
- Self-hosting (closed weights)
- Image generation
Research papers
- Algebraic World Model: From Exploration to Proofs29 Oct
- TreasuryQA Bench: Can AI Agents Work an Archive of Scanned Financial Records?28 Oct
- Investor's Last Exam: Can AI Agents Do Equity Investment Without Seeing the Future?28 Oct
- Agentic Distributional Regression: Conditional Density Estimation through Agent-Written Programs27 Oct
- The TEMPO network can be made independent of the system size27 Oct
- kw-bench: Breaking Frontier Harnesses on Real-World Tasks26 Oct
- VibeVideoBench:Evaluating Multimodal Agents on Code-Driven Video Generation26 Oct
- NumTrace: Auditing the Number Provenance of Financial Analyst Agents26 Oct
Social
Most watched and discussed · 90 days
- I Tested China's New Model vs Claude Opus 5.5 (Kimi K3.1)Nick Saraev · 77.2K views · 28 Oct
- Claude Opus 5.5 Brought NBA 2K27 to Roblox...Cole · 68.4K views · 28 Oct
- Unlimited & Free Claude Code + Claude Opus 5.5 for Free + Free Claude AI TutorialYo, Nadie · 61.0K views · 22 Oct
- Claude Opus 5.5 vs Gemini 4 Argon Make a Minecraft ModViral Echoes · 58.2K views · 21 Oct
- Claude Opus 5.5 is INSANE for Mobile App Animations (Apple’s Dynamic Island)Jason Lee · 51.9K views · 28 Oct
- Claude Opus 5.5 + GPT-6 Astra + Gemini 4 Argon for FREEYash Decode AI · 42.9K views · 25 Oct
- Claude Opus 5.5 & Fable 5.1 with Millions of Free Tokens (No Paid Plan)Think with MR · 41.4K views · 22 Oct
- Claude Opus 5.5 Makes Valorant in RobloxCole · 38.6K views · 30 Oct
- China's Free AI Kimi K3.1 Is Almost as Good as Claude Opus 5.5Aaghaz · 21.0K views · 28 Oct
- Gemini 4 Argon Лучшая Нейросеть в Мире? (vs Claude Opus 5.5)OBSION · 11.5K views · 29 Oct
- 30+ AI Models for FREE? Claude Opus 5.5 , GPT 6 Astra & Gemini 4 ArgonThe Tech King · 10.2K views · 25 Oct
- جربت الموديل الصيني (Kimi k3.1) الجديد ضد Claude opus 5.5Abdallah Khraisat · 9.6K views · 31 Oct
Same family
Claude Opus 5- 50.8
Alternatives
other vendors · similar capability- 54.5
- 55.1
- 56.9
- 59.4
Using Claude Opus 5.5 at volume?
Sign in to ask for a volume discount. You will continue in the OSMEXO dashboard.