Skip to content
Leaderboard / Anthropic

Claude Opus 4.1

Anthropic 2025-08-05 CLOSED Proprietary
Other settings:Thinking 16K, 2025-08-05 · 70
Verdict

Claude Opus 4.1 is a solid mid-pack model: #155 of 406 by SIscope score. Strongest at following instructions, weaker at math. On the pricey side: costs more than 100% of priced models.

SIscope score
56
#155 of 406
Human-vote rank
#124
of 419 · score 1419
Price per 1M tokens, blended
$30
$15 in · $75 out
Context window
200K
200,000 tokens

Shape

Each spoke is a category, 0 at the centre, 100 at the edge.

CodingAgentsno data yetReasoningMathChatInstructions

Category scores

Percentiles: 90 means better than about 90% of tested models.

Coding52
Agents & toolsNo data yet–
Reasoning & knowledge61
Math49
Chat quality71
Following instructions83

What goes into each category · Compare with DeepSeek V4.1 Flash

Human-vote score over time

Text arena, recent updates.

141014181425Oct 20: 1418.6 (rank #123)Oct 22: 1418.6 (rank #123)Oct 28: 1418.6 (rank #124)Oct 20Oct 28

Coding votes over time

WebDev arena, recent updates.

138513901395Oct 16: 1390.3 (rank #98)Oct 18: 1390.5 (rank #99)Oct 19: 1390.3 (rank #102)Oct 22: 1390.4 (rank #103)Oct 23: 1390.6 (rank #104)Oct 27: 1390.6 (rank #105)Oct 28: 1390.2 (rank #106)Oct 16Oct 28

Independent test runs

Run by Epoch AI on the same harness for every model. Bars show the share of questions answered correctly.

GPQA Diamond73%
PhD-level science multiple choice questions
Better than 45% of 333 tested models. Run Aug 5, 2025.
OTIS Mock AIME 2024-2540%
Competition maths problems in the style of AIME
Better than 29% of 313 tested models. Run Aug 5, 2025.
SWE-bench Verified73%
Fixing real GitHub issues in Python repositories
Better than 41% of 35 tested models. Run Feb 11, 2026.

Human-vote arenas

People compare two anonymous answers and pick the better one. Scores work like chess ratings.

ArenaRankScoreRange (95%)VotesUpdated
Text arena
Everyday chat, blind human votes
#124 of 41914191416–142277,213Oct 28, 2026
WebDev arena
Building web apps, blind human votes
#106 of 15213901379–14016,904Oct 28, 2026
Vision arena
Questions about images, blind human votes
Not on this board yet
Agent arena
Real multi-step agent sessions
Not on this board yet

Source: Arena (formerly LMArena) leaderboard dataset, CC BY 4.0. Fetched Nov 4, 2026.

Price

Input
$15
per 1M tokens
Output
$75
per 1M tokens
Blended
$30
3 in : 1 out, per 1M
Context window
200K
max tokens per request
Speed
Coming soon
our own tests

List price from the OpenRouter models API (anthropic/claude-opus-4.1), Nov 4, 2026. Providers can charge differently.