Skip to main content
leaderboard
.cn
Toggle Sidebar
Search models or organizations…
⌘K
Explore
Frontier Models
Historical Trends
Local Models
Compare Models
Ranking Studio
Library
Models
Pricing
Organizations
Benchmarks
Data sources
Settings
Toggle Sidebar
leaderboard
.cn
Loading page content
Feedback
Loading page content
Back to models
Claude Opus 4.8
Closed
Share
Add to comparison
Tool use · Reasoning
Overview
Pricing
Evaluations
Evaluation overview
Overall rank
#5
92.3 points · 180 models
Research rank
#3
99.2 points · 174 models
Data coverage
122 items
16 sources · updated through 2026-07-17
Domain performance
Coding
#4 / 160
96.8
Research
#3 / 174
99.2
Agents
#4 / 125
93.1
Vision
#16 / 65
87.9
Long context
#7 / 82
88.2
Evaluation highlights
All evaluations
SWE-bench Verified
Coding · Hy3 Hugging Face README
88.6%
1
/ 60
HLE (Humanity's Last Exam)
Reasoning · Hy3 Hugging Face README
49.8%
1
/ 57
Tool-Decathlon
Agents · GLM-5.2 Technical Report
59.9%
1
/ 21
Arena Vision Rating
Multimodal · Arena.ai Vision · Thinking mode
1,286.42Elo
7
/ 121
Arena Document Rating
Long context · Arena.ai Document · Thinking mode
1,476.64Elo
7
/ 28
IFEval
Instruction following · LongCat-2.0 Technical Report
86%
20
/ 30
Basic information
Context (input / output)
1M / 128K
Release date
2026-05
Input modalities
Text · Image
Output modalities
Text
Availability
Anthropic
Full pricing
Input
$5 /1M
Output
$25 /1M
Related links
Official site
Version relationships
Claude Opus 4 family
Claude Opus 4
2025-05 · Same series · Deprecated
Claude Opus 4.1
2025-08 · Same series · Deprecated
Claude Opus 4.5
2025-11 · Same series
Claude Opus 4.6
2026-02 · Same series
Claude Opus 4.7
2026-03 · Same series
Claude Opus 4.8
Current
2026-05 · Current model
Runtime modes
Thinking