Model
Model explorer

Claude Opus 4.1

CLOSED
Anthropic · Claude 4 family · released Aug 5, 2025

Incremental Opus 4 upgrade improving agentic coding, refactoring and reasoning benchmarks; drop-in replacement at same price.

ReasoningCodingVisionFunction callingTool useAgentic
1733.8
Elo · rank #104
Parameters
Undisclosed
Active params
—
Context
200K tokens
Architecture
Dense transformer, hybrid-reasoning (parameter count undisclosed)
License
Proprietary
Languages
—
API price (in/out)
$15 / $75
Modalities
text · vision
Benchmark results
Bar shows position within the tracked field; marker = field best
AIMEMath78.0%#85
best: GPT-5.2 · 100.0%
Arena EloHuman preference1447#10
best: Claude Fable 5 · 1505
GAIAAgents64.2%#3
best: MiniMax-M2 · 75.7%
GPQA DiamondReasoning80.9%#83
best: GPT-6 Astra · 96.0%
Humanity's Last ExamReasoning7.9%#107
best: Claude Opus 5 · 64.7%
LiveCodeBenchCoding64.6%#89
best: DeepSeek-V4-Pro (Think Max) · 93.5%
MMMLUKnowledge89.5%#7
best: Gemini 3.1 Pro · 92.6%
MMMUVision77.1%#30
best: Claude Fable 5 · 89.3%
OSWorldAgents44.4%#3
best: Seed 2.1 Pro · 78.8%
SWE-bench VerifiedCoding74.5%#46
best: Claude Opus 5 · 96.0%
tau-benchAgents82.4%#1
best: this model · 82.4%
Terminal-Bench 2.0Coding46.5%#65
best: Gemini 3.8 Flash · 89.4%
Terminal-BenchCoding43.3%#4
best: Claude Opus 4.5 (High) · 59.3%
Run it locally
Closed weights — available via API only. No local deployment.
Input / M tok
$15
Output / M tok
$75
Claude 4 family
Elo progression across releases
Claude Opus 4May 20251674.4Claude Sonnet 4May 20251566.9Claude Opus 4.1Aug 20251733.8Claude Sonnet 4.5Sep 20251938.8Claude Haiku 4.5Oct 20251556.0Claude Opus 4.5Nov 20252228.2Claude Opus 4.5 (High)Nov 20252261.5Claude Opus 4.5 (Medium)Nov 20252255.6Claude Opus 4.6Feb 20262433.7Claude Sonnet 4.6Feb 20262417.2Claude Opus 4.7Apr 20262674.5Claude Opus 4.8May 20262756.0
Resources
DOCSplatform.claude.comAPIAvailable via Anthropic API
API price $15/$75 · each benchmark row carries its own source badge (see methodology)