Claude Opus 4.6
anthropic/claude-opus-4.6- Anthropic
- Family · Claude Opus 4
- Released 5 Feb 2026
- Text · Image · File → Text
- Reasoning · adjustable
- Tools · JSON · Vision
Not on OSMEXO yet
Spec & list price
Context1M
Max output128K
Input / 1M$5.00
Output / 1M$25.005× input
Scores · standing
Intelligence31.9#60/489
Usage
Tokens / day13.6B▲ 22% 7d
Usage rank#81
Requests / day310.6K
Tool calls / req0.24
Per request · 30 days
Avg input33.7K2.3× median
Avg output689about median
Cache hit65%about median
Reasoning share14%0.3× median
Tool-call failures0.6%0.3× median
Consumption
Peak 18.2B 29 OctLow 9.6B 24 OctRank today #81
News & trends
5 mentions this week ▼ 44%this model · videos, posts, articles · 30 days
Trending · 14 days
- Abril de 2026: un agente de programación (Cursor con Claude Opus 4.6) trabajaba en el entorno de…1.2K views
- They said Hunter x Hunter is a kids anime - Claude Opus 4.61.1K views
- I spent a couple of days benchmarking local Qwen models against Claude Opus 4.6 — here are the results100 upvotes
News · latest from Anthropic
- Claude Opus 5.5 Can Draw Permit-Level Architectural Drawingsstephenjgallagher.substack.com · 2 Nov
- GPT-6.1 Terra vs. Claude Opus 5.5: Cheaper and faster in three tests at maxthenewstack.io · 2 Nov
- Claude Opus 5.5 and Claude Sonnet 5.5 are now available in Google AntigravityGoogle Cloud · 20 Oct
- GLM-5.3 vs. Claude Fable 5 on DeepSWE: Cost, Coding, and RoutingTogether AI · 21 Aug
Benchmarks
bar = share of the top score · Benchmark data: Artificial AnalysisSpeed · median across 5 providers, latest day
Output speed#284/33436 tok/s1.8× slower than median
Fastest provider38 tok/sbest of 5 providers
First token#2501.83 s1.8× longer than median
#13 cheapest of 14 with a similar score
Value
Intelligence Index
Blended price per 1M tokens (log)
Family progress
Intelligence Index
- Family releases41.8
Technical details
- Reasoning
- on · high
- Tokenizer
- Claude
- Moderation
- moderated
Reasoning effort
- max
- high · default
- medium
- low
- none
Request parameters 14
- tools
- tool_choice
- response_format
- structured_outputs
- reasoning
- reasoning_effort
- include_reasoning
- temperature
- top_p
- top_k
+4 more
- max_tokens
- max_completion_tokens
- stop
- verbosity
When to use
Best for
- Hard science question answering
- Tool-calling agents and long document reasoning
- Large-codebase analysis with big outputs
- Image and file analysis
Think twice for
- High-volume simple chat (cheaper options exist)
- Self-hosting (closed weights)
- New builds where newer Opus releases fit better
Research papers
- Vendster: An Evaluation Arena to Investigate How Forecasting Tool Quality Affects LLM Agent Behaviors in Buyer-Vendor Negotiation28 Oct
- PRISM-BENCH: Diagnosing Grounded and Faithful Multimodal Reasoning Beyond Final-Answer Accuracy27 Oct
- TIDE: Tree-Induced Fine-Grained Diagnostic Evaluation from Model Behaviors21 Oct
- BIRD-RL: Scaling Agentic Reinforcement Learning over Stateful Data-Centric Environments21 Oct
- Exposing Vulnerability Monoculture Through Systemic Adversarial Robustness19 Oct
- Same Simulation, Different Evidence: Where Retrieved Tool Output Costs LLM Agents Accuracy19 Oct
- Can LLMs Fix It Without Code? Toward Automated Verification of No-Code Bug Fixes8 Oct
- The Pushback Paradox: A Two-Probe Diagnostic for Language Model Compliance5 Oct
Same family
Claude Opus 4- 41.8
- 40.7
- 29.1
Alternatives
other vendors · similar capability- 29.7
- 27.2
- 28.0
- C33.9
Want Claude Opus 4.6 on OSMEXO?
Sign in to request it. Requests are reviewed; a request does not guarantee availability.






