Claude Haiku 5.5 Benchmark Details
Claude Haiku 5.5 currently shows benchmark results led by HLE (18 / 244, score 57.40), APEX-SWE (9 / 73, score 55.10), CursorBench 4.0 (11 / 75, score 48.40). This page also tracks comparisons against 1 predecessor or same-series models.
Benchmark Results
Benchmark Results
Abstract Generalization
11 evaluationsCross-industry Work
2 evaluationsOffice & Business
6 evaluationsMathematics
2 evaluationsRepository Engineering
2 evaluationsDocuments & Charts
2 evaluationsAgentic Development
6 evaluationsCode Generation & Editing
2 evaluationsCapability Indices
5 evaluationsVersion History
How each version of the Claude Haiku 5.5 series stacks up on benchmark tests
4 benchmarks with comparable scores. Each model shows its best score; mode label is displayed below.· Click a row to view its trend chart.
| Benchmark | Claude Haiku 5.5Current | Haiku 4.5 |
|---|---|---|
89.00Thinking Level · High | 47.67Extended Thinking | |
61.25Thinking Level · Extra High | 4.50Extended Thinking | |
57.40Thinking Level · High ｜ Tools | 9.70Extended Thinking | |
75.64Standard Mode ｜ Tools | 46.04Standard Mode ｜ Tools |
Single-Benchmark Version Trend
Viewing: ARC-AGI-1 · Abstract Generalization
Standard API Pricing Across the Claude Haiku 5.5 Series
Shows standard text input and output pricing side by side for each model. If extended-context pricing exists, the chart keeps the base rate and explains the threshold below.
Source: DataLearnerAI. Standard text prices shown here use the default supplier. · USD / 1M tokens
When a context threshold exists, the charted base price only applies within these limits:
| Model | Supplier | Standard input | Standard output | Base price applies to |
|---|---|---|---|---|
Claude Haiku 5.5 | Anthropic | $0.1 / 1M tokens | $0.5 / 1M tokens | <= 100000 |
Haiku 4.5 | Anthropic | $1 / 1M tokens | $5 / 1M tokens | — |