modelgrep
A

Anthropic: Claude Opus 4.6

anthropic/claude-opus-4.6

12th smartest of 179ReasoningToolsJSONVision
Use via OpenRouter ↗
Intelligence
52.9
12th of 179
Design Elo
1358
3D
Speed
41
202nd fastest
Latency
1.5s
first token
Input price
$5.00
277th cheapest
Context
1M
128K max out

How it compares

Smarter than93%
of all ranked models
Faster than34%
of all ranked models
Cheaper than7%
of all ranked models

Overview

Opus 4.6 is Anthropic’s strongest model for coding and long-running professional tasks. It is built for agents that operate across entire workflows rather than single prompts, making it especially effective...

Benchmarks

independent · via OpenRouter
Artificial Analysis94th percentile
Intelligence Index
52.9
Coding Index
48.1
Agentic Index
67.6
GPQA Diamond
90%
Humanity's Last Exam
37%
SciCode
52%
Tau²-Bench (agentic)
92%
Design Arena · Elo16,618 tournaments
3D
1358
Game Dev
1339
codecategories
1338
UI Component
1337
Website
1332
Data Viz
1317
asciiart
1303
fullstack
1299
webapps
1282
svg
1279
mobileapps
1269
androidnative
1221

Providers & pricing (6)

ProviderIn $/MOut $/MUptime
Google$5.00$25.00
Amazon Bedrock$5.00$25.0099.9%
Azure$5.00$25.00
Anthropic$5.00$25.00100%
Google$5.00$25.00100%
Anthropic$5.00$25.00

Specifications

Context window1M
Max output128K
Knowledge cutoff
Input modalitiestext, image, file
Output modalitiestext
Prompt caching
Cache read price$0.500/M
ModeratedNo

Claude Opus 4.6 FAQ

How much does Claude Opus 4.6 cost?

Claude Opus 4.6 costs $5.00 per million input tokens and $25.00 per million output tokens via OpenRouter, making it 277th cheapest of 298 paid models.

How smart is Claude Opus 4.6?

Claude Opus 4.6 scores 52.9 on the Artificial Analysis Intelligence Index, ranking 12th of 179 benchmarked models, with a GPQA Diamond score of 90%.

How fast is Claude Opus 4.6?

Claude Opus 4.6 generates around 41 tokens per second with 1.5s time-to-first-token (p50), the 202nd fastest tracked model.

What is Claude Opus 4.6's context window?

Claude Opus 4.6 supports a 1M-token context window and can output up to 128K tokens. It accepts text, image, file input.

Compare head-to-head