Claude Opus 4.8
ClaudeFlagship
ThinkingTool UseVisionStructured Output
About this model
Claude Opus 4.8 is Anthropic's most capable generally available model in the Opus family. It supports text, image, and file inputs with text output, with reasoning support and a 1M-token...
Performance Tier
Flagship
Claude Opus 4.8 is a flagship model from Claude : the most capable in their lineup.
Best-in-class model from this provider. Highest performance across benchmarks, ideal for demanding tasks.
Pricing
This model is included in Elosia plansPremium
Highest cost level. A long conversation can quickly consume your monthly cap.
| Type | per 1M tokens |
|---|---|
| Input (prompt) | $5.00 |
| Output (completion) | $25.00 |
| Cache read | $0.500 |
| Cache write | $6.25 |
Capabilities
Context Length1.0M
Max Output Tokens128K
TokenizerClaude
Inputtext, image, file
Outputtext
Release DateMay 27, 2026
Benchmarks
General Intelligence
MMLU
Not reported
GPQA Diamond
93.6%
Mathematics
MATH-500
Not reported
Programming
HumanEval
Not reported
SWE-bench Verified
88.6%
Reasoning
IFEval
Not reported
Humanity's Last Exam
49.8%
Agentic
SWE-bench Pro
69.2%
Recommended Use Cases
CodingAnalysisResearchCreative Writing
Strengths
- Best-in-class real-world software engineering (SWE-bench Verified 88.6%, SWE-bench Pro 69.2%)
- State-of-the-art graduate-level scientific reasoning (GPQA Diamond 93.6%)
- Adaptive thinking — reasons only when the task needs it, cutting wasted thinking tokens
- Around 4× less likely than Opus 4.7 to let flaws in its own code pass unremarked
- Robust long-horizon agentic coding with 1M-token context and improved compaction recovery
Limitations
- Premium pricing ($5 / $25 per million input / output tokens)
- MMLU, MATH-500, HumanEval, IFEval and ARC-AGI not reported — Anthropic considers them saturated
- Higher time-to-first-token; slower than Sonnet/Haiku for latency-sensitive queries
Resources
This model may use your data for training
