picksbymodel Get API access →
Anthropic · 28 models on OpenRouter

Best Anthropic Models (2026)

Claude Opus, Sonnet, and Haiku ranked by task. Anthropic's Claude models consistently lead on reasoning, code quality, and safety benchmarks. Ideal for agentic workflows and long-context tasks.

Top Anthropic models by benchmark

#ModelScoreIn / 1MOut / 1MContext
1 Anthropic: Claude Opus 5.5 (batch) 100 $2.00 $10.00 1,000,000 Details
2 Anthropic: Claude Opus 5 (batch) 100 $2.50 $12.50 1,000,000 Details
3 Anthropic: Claude Sonnet 5 100 $2.00 $10.00 1,000,000 Details
4 Anthropic: Claude Sonnet 5 (batch) 100 $1.00 $5.00 1,000,000 Details
5 Anthropic: Claude Opus 4.8 (batch) 100 $2.50 $12.50 1,000,000 Details
6 Anthropic: Claude Opus 4.7 (batch) 100 $2.50 $12.50 1,000,000 Details
7 Anthropic: Claude Sonnet 4.6 100 $3.00 $15.00 1,000,000 Details
8 Anthropic: Claude Sonnet 4.6 (batch) 100 $1.50 $7.50 1,000,000 Details
9 Anthropic: Claude Opus 4.6 (batch) 100 $2.50 $12.50 1,000,000 Details
10 Anthropic: Claude Opus 4.5 (batch) 100 $2.50 $12.50 200,000 Details
11 Anthropic: Claude Haiku 4.5 100 $1.00 $5.00 200,000 Details
12 Anthropic: Claude Haiku 4.5 (batch) 100 $0.50 $2.50 200,000 Details
13 Anthropic: Claude Sonnet 4.5 100 $3.00 $15.00 1,000,000 Details
14 Anthropic: Claude Sonnet 4.5 (batch) 100 $1.50 $7.50 1,000,000 Details
15 Anthropic: Claude Opus 5.5 98 $4.00 $20.00 1,000,000 Details

All Anthropic models (28 total)

Anthropic: Claude Opus 5.5 (batch)
$2.000/M in
1000000ctx · score 100
Anthropic: Claude Opus 5 (batch)
$2.500/M in
1000000ctx · score 100
Anthropic: Claude Sonnet 5
$2.000/M in
1000000ctx · score 100
Anthropic: Claude Sonnet 5 (batch)
$1.000/M in
1000000ctx · score 100
Anthropic: Claude Opus 4.8 (batch)
$2.500/M in
1000000ctx · score 100
Anthropic: Claude Opus 4.7 (batch)
$2.500/M in
1000000ctx · score 100
Anthropic: Claude Sonnet 4.6
$3.000/M in
1000000ctx · score 100
Anthropic: Claude Sonnet 4.6 (batch)
$1.500/M in
1000000ctx · score 100
Anthropic: Claude Opus 4.6 (batch)
$2.500/M in
1000000ctx · score 100
Anthropic: Claude Opus 4.5 (batch)
$2.500/M in
200000ctx · score 100
Anthropic: Claude Haiku 4.5
$1.000/M in
200000ctx · score 100
Anthropic: Claude Haiku 4.5 (batch)
$0.500/M in
200000ctx · score 100
Anthropic: Claude Sonnet 4.5
$3.000/M in
1000000ctx · score 100
Anthropic: Claude Sonnet 4.5 (batch)
$1.500/M in
1000000ctx · score 100
Anthropic: Claude Opus 5.5
$4.000/M in
1000000ctx · score 98
Anthropic: Claude Sonnet 4
$3.000/M in
200000ctx · score 96
Anthropic: Claude Fable 5.1 (batch)
$5.000/M in
1000000ctx · score 95
Anthropic: Claude Opus 5
$5.000/M in
1000000ctx · score 95
Anthropic: Claude Fable 5 (batch)
$5.000/M in
1000000ctx · score 95
Anthropic: Claude Opus 4.8
$5.000/M in
1000000ctx · score 95
Anthropic: Claude Opus 4.7
$5.000/M in
1000000ctx · score 95
Anthropic: Claude Opus 4.6
$5.000/M in
1000000ctx · score 95
Anthropic: Claude Opus 4.5
$5.000/M in
200000ctx · score 95
Anthropic: Claude 3 Haiku
$0.250/M in
200000ctx · score 94
Anthropic: Claude Opus 4.1 (batch)
$7.500/M in
200000ctx · score 88
Anthropic: Claude Fable 5.1
$10.000/M in
1000000ctx · score 85
Anthropic: Claude Fable 5
$10.000/M in
1000000ctx · score 85
Anthropic: Claude Opus 4.1
$15.000/M in
200000ctx · score 80

Anthropic strengths by task

Code

Best for SQL Generation

Anthropic ranks #1
Code

Best for Code Review

Anthropic ranks #1
Code

Best for Code Completion

Anthropic ranks #1
Code

Best for Code Refactoring

Anthropic ranks #1
Code

Best for Bug Fixing

Anthropic ranks #1
Code

Best for Unit Test Generation

Anthropic ranks #1

The Model Movers Report

One email every Friday, built from this site's own rankings: the current top five by benchmark score, every model released in the last seven days, and one note worked out from that week's numbers. You can unsubscribe from any issue with one click.