Skip to content
Leaderboard / Anthropic

Claude Opus 4.5

Anthropic 2025-11-01 CLOSED Proprietary
Other settings:High, 32K, 2025-11-01 · 73
Verdict

Claude Opus 4.5 is a solid mid-pack model: #84 of 406 by SIscope score. Strongest at following instructions, weaker at math. On the pricey side: costs more than 94% of priced models.

SIscope score
70
#84 of 406
Human-vote rank
#52
of 419 · score 1451
Price per 1M tokens, blended
$10
$5 in · $25 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.

Coding72
Agents & toolsNo data yet–
Reasoning & knowledge74
Math59
Chat quality88
Following instructions93

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

Human-vote score over time

Text arena, recent updates.

144514501455Oct 20: 1451 (rank #50)Oct 22: 1450.9 (rank #50)Oct 28: 1450.9 (rank #52)Oct 20Oct 28

Coding votes over time

WebDev arena, recent updates.

146514701475Oct 16: 1469.4 (rank #67)Oct 18: 1469.4 (rank #68)Oct 19: 1469.4 (rank #70)Oct 22: 1469.5 (rank #71)Oct 23: 1469.5 (rank #72)Oct 27: 1469.3 (rank #73)Oct 28: 1469.2 (rank #74)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 Diamond81%
PhD-level science multiple choice questions
Better than 58% of 333 tested models. Run Nov 24, 2025.
OTIS Mock AIME 2024-2548%
Competition maths problems in the style of AIME
Better than 34% of 313 tested models. Run Nov 24, 2025.
SWE-bench Verified77%
Fixing real GitHub issues in Python repositories
Better than 71% of 35 tested models. Run Feb 5, 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
#52 of 41914511448–145472,825Oct 28, 2026
WebDev arena
Building web apps, blind human votes
#74 of 15214691462–147613,362Oct 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
$5
per 1M tokens
Output
$25
per 1M tokens
Blended
$10
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-4.5), Nov 4, 2026. Providers can charge differently.