Model comparison
Claude Haiku 5.5 vs Claude Sonnet 5
Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 49.5 on the Noometry Index. Claude Haiku 5.5 costs 20× less per token, which makes it the better buy when Claude Sonnet 5's lead doesn't matter for your workload.
Last verified . 8 shared benchmarks.
Summary
- They share 8 benchmarks with published results for both. Claude Haiku 5.5 scores higher in 2 categories and Claude Sonnet 5 in 3 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where Claude Sonnet 5 leads 49.1 to 35.2.
- The biggest single-benchmark swing is FrontierMath Tier 4: 46.3% for Claude Haiku 5.5 and 29.3% for Claude Sonnet 5.
- Claude Haiku 5.5 is cheaper at $0.10 / $0.50 per million input/output tokens, against $2 / $10 for Claude Sonnet 5.
Side by side
| Claude Haiku 5.5 | Claude Sonnet 5 | |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 49.5 | 54.6 |
| Released | 2026-10-07 | 2026-06-29 |
| Weights | Proprietary | Proprietary |
| Context window | 1M | 1M |
| Max output | 128K | 128K |
| Input $ / M tokens | $0.10 | $2 |
| Output $ / M tokens | $0.50 | $10 |
| Results tracked | 9 | 51 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Sonnet 5 leads
Claude Haiku 5.5: 49.2 (#50), Claude Sonnet 5: 55.5 (#26)
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 5 |
|---|---|---|
| LMArena WebDev | 1587 | 1541 |
| DeepSWE | — | 53.8% |
| FrontierCode | — | 42.7% |
| CursorBench | — | 34.1% |
| SciCode | — | 54.3% |
| GSO | — | 37.3% |
| WeirdML | — | 68.8% |
| LMArena Coding | — | 1483 |
| ALE-Bench | — | 1,463 |
Agentic & Tool Use Not comparable
Claude Haiku 5.5: —, Claude Sonnet 5: 42.8 (#18)
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 5 |
|---|---|---|
| APEX-Agents | — | 54.5% |
| GBAEval | — | 65.3% |
| LMArena Search | — | 1194 |
| Vending-Bench 2 | — | 6,378 |
Reasoning Claude Sonnet 5 leads
Claude Haiku 5.5: 35.2 (#69), Claude Sonnet 5: 49.1 (#39)
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 5 |
|---|---|---|
| NYT Connections (extended) | 65.7% | 75.1% |
| Mystery Game Puzzles | 30% | 35% |
| SimpleBench | — | 60.6% |
| CritPt | — | 16.9% |
| Chess Puzzles | — | 35% |
| LMArena Hard Prompts | — | 1461 |
| DTBench | — | 92.5% |
| LMCA | — | 50% |
| Surface Evolver Bench | — | 60% |
| Bench to the Future 3 | — | 0.14 |
| Epoch Capabilities Index | — | 156.21 |
| ForecastBench | — | 61.1 |
Math Claude Haiku 5.5 leads
Claude Haiku 5.5: 73.6 (#18), Claude Sonnet 5: 66.2 (#27)
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 5 |
|---|---|---|
| FrontierMath (Tiers 1-3) | 75.1% | 65.6% |
| FrontierMath Tier 4 | 46.3% | 29.3% |
| OTIS Mock AIME 2024-2025 | 98.9% | 94.7% |
| ProofBench | — | 77% |
| LMArena Math | — | 1467 |
Knowledge Claude Sonnet 5 leads
Claude Haiku 5.5: 50.8 (#70), Claude Sonnet 5: 55.6 (#47)
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 5 |
|---|---|---|
| GPQA Diamond | 89.6% | 90.5% |
| SimpleQA Verified | 23.8% | 33.7% |
| LMArena Expert | — | 1490 |
Multimodal Claude Haiku 5.5 leads
Claude Haiku 5.5: 43.7 (#21), Claude Sonnet 5: 42.4 (#31)
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 5 |
|---|---|---|
| LMArena Vision | — | 1274 |
| Blueprint-Bench 2 | — | 24.9% |
| Furniture Assembly | 47.5% | — |
| LMArena Document | — | 1466 |
Multilingual Not comparable
Claude Haiku 5.5: —, Claude Sonnet 5: 53.8 (#55)
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 5 |
|---|---|---|
| LMArena Non-English | — | 1431 |
| LMArena Chinese | — | 1477 |
| LMArena French | — | 1460 |
| LMArena German | — | 1440 |
| LMArena Japanese | — | 1422 |
| LMArena Korean | — | 1411 |
| LMArena Russian | — | 1451 |
| LMArena Spanish | — | 1437 |
Instruction Following Not comparable
Claude Haiku 5.5: —, Claude Sonnet 5: 76.3 (#41)
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 5 |
|---|---|---|
| LMArena Instruction Following | — | 1452 |
Long Context Not comparable
Claude Haiku 5.5: —, Claude Sonnet 5: 44.8 (#55)
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 5 |
|---|---|---|
| LMArena Longer Query | — | 1463 |
Writing & Preference Not comparable
Claude Haiku 5.5: —, Claude Sonnet 5: 69.2 (#25)
| Benchmark | Claude Haiku 5.5 | Claude Sonnet 5 |
|---|---|---|
| LMArena Text | — | 1442 |
| LMArena Creative Writing | — | 1416 |
| EQ-Bench Creative Writing | — | 1794 |
| EQ-Bench 4 | — | 1236 |
| LMArena Multi-Turn | — | 1454 |
Frequently asked questions
Is Claude Haiku 5.5 better than Claude Sonnet 5?
Claude Sonnet 5 is the stronger model overall, scoring 54.6 to 49.5 on the Noometry Index. Claude Haiku 5.5 costs 20× less per token, which makes it the better buy when Claude Sonnet 5's lead doesn't matter for your workload.
Which is cheaper, Claude Haiku 5.5 or Claude Sonnet 5?
Claude Haiku 5.5 is cheaper. It lists at $0.10 per million input tokens and $0.50 per million output tokens; Claude Sonnet 5 lists at $2 and $10.
Is Claude Haiku 5.5 or Claude Sonnet 5 better for coding?
Claude Sonnet 5 scores higher on coding benchmarks: 55.5 versus 49.2 in the Noometry coding category.
Which has the bigger context window?
Both accept 1M tokens.
How many benchmarks do Claude Haiku 5.5 and Claude Sonnet 5 share?
8 benchmarks have published results for both models. Claude Haiku 5.5 has 9 scored results on Noometry and Claude Sonnet 5 has 51.