# Claude 4 Sonnet vs Claude Haiku 4.5

> Side-by-side benchmark comparison for Claude 4 Sonnet and Claude Haiku 4.5 across agentic, coding, multimodal, knowledge, reasoning, multilingual, and math tasks.

- Canonical page: https://benchlm.ai/compare/claude-4-sonnet-vs-claude-haiku-4-5
- Last updated: September 29, 2026

- Shared sourced benchmarks: 2
- HTML indexing: indexable
- Ranking lane: BenchAlign v5.7

## Quick Verdict

Evidence is too thin for a benchmark verdict: Claude 4 Sonnet and Claude Haiku 4.5 share only 2 sourced benchmark rows.

## Summary

- Only 2 shared sourced benchmark rows are available. That is not enough for a broad capability verdict, so this mirror does not name an overall winner.
- Claude Haiku 4.5 has the lower published output-token price.

## Model Snapshot

| Property | Claude 4 Sonnet | Claude Haiku 4.5 |
|----------|----------|----------|
| Creator | Anthropic | Anthropic |
| Type | Proprietary | Proprietary |
| Reasoning | Non-Reasoning | Non-Reasoning |
| Context | 200K | 200K |
| Independent public score (not head-to-head) | 35.92 | 42.47 |
| Benchmarks Covered | 3 | 10 |
| Pricing (input/output) | $3.00 / $15.00 | $1.00 / $5.00 |

## Category Breakdown

### Agentic

- Winner: Insufficient shared evidence
- Claude 4 Sonnet public-lane score: Coming soon
- Claude Haiku 4.5 public-lane score: 22 (Supported · #96/117)

| Benchmark | Claude 4 Sonnet | Claude Haiku 4.5 | Winner |
|-----------|-----------|-----------|--------|
| Gert Labs | 39.66% | Coming soon | Coming soon |
| JobBench | 18.4% | 16.0% | Claude 4 Sonnet |
| Terminal-Bench 2.1 (Vals) | Coming soon | 43.8% | Coming soon |

### Coding

- Winner: Insufficient shared evidence
- Claude 4 Sonnet public-lane score: Coming soon
- Claude Haiku 4.5 public-lane score: 19.6 (Supported · #124/143)

| Benchmark | Claude 4 Sonnet | Claude Haiku 4.5 | Winner |
|-----------|-----------|-----------|--------|
| SWE-bench Verified | 72.7% | 73.3% | Claude Haiku 4.5 |
| VulcanBench v3 | Coming soon | 76.2% | Coming soon |
| LiveCodeBench (Vals) | Coming soon | 41.2% | Coming soon |
| SWE-bench (Vals) | Coming soon | 66.6% | Coming soon |

### Reasoning

- Winner: Insufficient shared evidence
- Claude 4 Sonnet public-lane score: 52.5 (Unranked · 2 rankable rows)
- Claude Haiku 4.5 public-lane score: Coming soon

### Multimodal & Grounded

- Winner: Insufficient shared evidence
- Claude 4 Sonnet public-lane score: 53 (Unranked · 1 rankable row)
- Claude Haiku 4.5 public-lane score: Coming soon

### Knowledge

- Winner: Insufficient shared evidence
- Claude 4 Sonnet public-lane score: 37 (Estimated · #108/169)
- Claude Haiku 4.5 public-lane score: 34.8 (Estimated · #119/169)

| Benchmark | Claude 4 Sonnet | Claude Haiku 4.5 | Winner |
|-----------|-----------|-----------|--------|
| GPQA Diamond (Vals) | Coming soon | 72.2% | Coming soon |
| MMLU-Pro (Vals) | Coming soon | 78.7% | Coming soon |

### Instruction Following

- Winner: Insufficient shared evidence
- Claude 4 Sonnet public-lane score: 52 (#78/124)
- Claude Haiku 4.5 public-lane score: Coming soon

### Mathematics

- Winner: Insufficient shared evidence
- Claude 4 Sonnet public-lane score: Coming soon
- Claude Haiku 4.5 public-lane score: 28.8 (Unranked · 2 rankable rows)

| Benchmark | Claude 4 Sonnet | Claude Haiku 4.5 | Winner |
|-----------|-----------|-----------|--------|
| FrontierMath v2 (Tiers 1-3) | Coming soon | 5.903% | Coming soon |
| FrontierMath v2 (Tier 4) | Coming soon | 2.083% | Coming soon |

## FAQ

### Is there enough evidence to pick between Claude 4 Sonnet and Claude Haiku 4.5?

Not yet. The models share only 2 sourced benchmark rows, so this mirror reports the available measurements and metadata without naming a broad winner.

### What can I compare safely on this page?

Published pricing, context windows, model type, and individual shared benchmark rows are directly comparable. Independent overall scores and non-overlapping benchmark coverage are labeled separately and should not be read as a controlled head-to-head result.

## Related Comparisons

- [Claude 4 Sonnet vs Claude Haiku 4.5 Thinking](/compare/claude-4-sonnet-vs-claude-haiku-4-5-thinking)
- [Claude Haiku 4.5 vs Claude Haiku 4.5 Thinking](/compare/claude-haiku-4-5-vs-claude-haiku-4-5-thinking)
- [Claude 4 Sonnet vs GPT-6 Astra](/compare/claude-4-sonnet-vs-gpt-6-astra)
- [Claude Haiku 4.5 vs GPT-6 Astra](/compare/claude-haiku-4-5-vs-gpt-6-astra)

## Explore More

- [Claude 4 Sonnet profile](/models/claude-4-sonnet)
- [Claude Haiku 4.5 profile](/models/claude-haiku-4-5)
- [Compare Pricing](/llm-pricing)
- [Alternative Finder](/tools/alternative-finder)
- [LLM Selector](/tools/llm-selector)
- [Overall Rankings](/best/overall)