Model comparison
Claude 3.5 Haiku vs Claude Haiku 4.5
Claude Haiku 4.5 is the stronger model overall, scoring 39.5 to 29.2 on the Noometry Index.
Last verified . 32 shared benchmarks.
Summary
- They share 32 benchmarks with published results for both. Claude 3.5 Haiku scores higher in 1 category and Claude Haiku 4.5 in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Claude Haiku 4.5 leads 44.9 to 14.7.
- The biggest single-benchmark swing is OTIS Mock AIME 2024-2025: 4.3% for Claude 3.5 Haiku and 66.7% for Claude Haiku 4.5.
Side by side
| Claude 3.5 Haiku | Claude Haiku 4.5 | |
|---|---|---|
| Provider | Anthropic | Anthropic |
| Noometry Index | 29.2 | 39.5 |
| Released | 2024-10-22 | 2025-10-15 |
| Weights | Proprietary | Proprietary |
| Context window | — | 200K |
| Max output | — | 64K |
| Input $ / M tokens | — | $1 |
| Output $ / M tokens | — | $5 |
| Results tracked | 49 | 53 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Claude Haiku 4.5 leads
Claude 3.5 Haiku: 32.9 (#265), Claude Haiku 4.5: 44.0 (#78)
| Benchmark | Claude 3.5 Haiku | Claude Haiku 4.5 |
|---|---|---|
| SciCode | 27.4% | 43.3% |
| WeirdML | 30.7% | 45.4% |
| LMArena Coding | 1286 | 1453 |
| SWE-bench Verified (bash only) | — | 66.6% |
| Aider Polyglot | 28% | — |
| LMArena WebDev | — | 1330 |
| SWE-bench Multilingual | — | 64.7% |
| BigCodeBench Instruct | 46.1% | — |
| LiveBench Coding | 51.4% | — |
| BigCodeBench Complete | 59% | — |
| CadEval | 32% | — |
| ALE-Bench | — | 653.48 |
Agentic & Tool Use Claude Haiku 4.5 leads
Claude 3.5 Haiku: 28.0 (#95), Claude Haiku 4.5: 33.6 (#52)
| Benchmark | Claude 3.5 Haiku | Claude Haiku 4.5 |
|---|---|---|
| BALROG | 19.3% | 31.2% |
| Terminal-Bench | — | 35.5% |
| Berkeley Function Calling Leaderboard | — | 68.7% |
| DeepResearch Bench | — | 45.5% |
| ExploitBench | — | 13.7% |
| Vending-Bench 2 | — | 458.89 |
Reasoning Claude 3.5 Haiku leads
Claude 3.5 Haiku: 17.7 (#290), Claude Haiku 4.5: 15.1 (#320)
| Benchmark | Claude 3.5 Haiku | Claude Haiku 4.5 |
|---|---|---|
| CritPt | 0% | 0% |
| LMArena Hard Prompts | 1251 | 1420 |
| DTBench | 56.7% | 73.6% |
| Epoch Capabilities Index | 127.15 | 142.41 |
| ARC-AGI-2 | — | 4% |
| NYT Connections (extended) | — | 14.3% |
| ARC-AGI-1 | — | 47.7% |
| Chess Puzzles | — | 8% |
| LiveBench Reasoning | 28.1% | — |
| LiveBench Data Analysis | 48.5% | — |
| LMCA | — | 30.9% |
| ForecastBench | — | 61.4 |
| LiveBench | 43.5% | — |
Math Claude Haiku 4.5 leads
Claude 3.5 Haiku: 14.7 (#300), Claude Haiku 4.5: 44.9 (#78)
| Benchmark | Claude 3.5 Haiku | Claude Haiku 4.5 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.3% | 66.7% |
| Omni-MATH | 22.4% | 56.1% |
| LMArena Math | 1244 | 1396 |
| MATH Level 5 | 46.4% | 96.4% |
| FrontierMath (Feb 2025 set) | 0.3% | 5.9% |
| LiveBench Math | 35.5% | — |
| FrontierMath Tier 4 (v1) | — | 2.1% |
Knowledge Claude Haiku 4.5 leads
Claude 3.5 Haiku: 18.7 (#281), Claude Haiku 4.5: 37.7 (#153)
| Benchmark | Claude 3.5 Haiku | Claude Haiku 4.5 |
|---|---|---|
| GPQA Diamond | 38.1% | 71.2% |
| MMLU-Pro | 60.5% | 77.7% |
| GPQA (HELM) | 36.3% | 60.5% |
| LMArena Expert | 1208 | 1442 |
| SimpleQA Verified | — | 13.2% |
| Confabulations | 36.7% | — |
| Vectara Hallucination Rate | — | 9.8% |
| MMLU | 74.3% | — |
Multimodal Too close to call
Claude 3.5 Haiku: 26.8 (#117), Claude Haiku 4.5: 26.8 (#118)
| Benchmark | Claude 3.5 Haiku | Claude Haiku 4.5 |
|---|---|---|
| LMArena Vision | 1092 | — |
| GeoBench | 34% | — |
| Blueprint-Bench 2 | — | 0% |
| LMArena Document | — | 1420 |
Multilingual Claude Haiku 4.5 leads
Claude 3.5 Haiku: 40.0 (#218), Claude Haiku 4.5: 49.9 (#129)
| Benchmark | Claude 3.5 Haiku | Claude Haiku 4.5 |
|---|---|---|
| LMArena Non-English | 1238 | 1377 |
| LMArena Chinese | 1229 | 1417 |
| LMArena French | 1264 | 1408 |
| LMArena German | 1237 | 1375 |
| LMArena Japanese | 1175 | 1339 |
| LMArena Korean | 1173 | 1347 |
| LMArena Russian | 1253 | 1381 |
| LMArena Spanish | 1261 | 1420 |
Instruction Following Claude Haiku 4.5 leads
Claude 3.5 Haiku: 62.9 (#234), Claude Haiku 4.5: 71.4 (#149)
| Benchmark | Claude 3.5 Haiku | Claude Haiku 4.5 |
|---|---|---|
| IFEval | 79.2% | 80.1% |
| LMArena Instruction Following | 1241 | 1414 |
| LiveBench Instruction Following | 61.9% | — |
Long Context Claude Haiku 4.5 leads
Claude 3.5 Haiku: 38.3 (#200), Claude Haiku 4.5: 43.6 (#92)
| Benchmark | Claude 3.5 Haiku | Claude Haiku 4.5 |
|---|---|---|
| LMArena Longer Query | 1261 | 1427 |
Writing & Preference Claude Haiku 4.5 leads
Claude 3.5 Haiku: 42.7 (#234), Claude Haiku 4.5: 57.9 (#123)
| Benchmark | Claude 3.5 Haiku | Claude Haiku 4.5 |
|---|---|---|
| LMArena Text | 1255 | 1396 |
| LMArena Creative Writing | 1233 | 1372 |
| WildBench | 76% | 83.9% |
| LMArena Multi-Turn | 1265 | 1409 |
| Short-Story Creative Writing | 73.5% | — |
| EQ-Bench Creative Writing | 1146 | — |
| EQ-Bench 4 | — | 1064 |
| LiveBench Language | 35.4% | — |
Frequently asked questions
Is Claude 3.5 Haiku better than Claude Haiku 4.5?
Claude Haiku 4.5 is the stronger model overall, scoring 39.5 to 29.2 on the Noometry Index.
Is Claude 3.5 Haiku or Claude Haiku 4.5 better for coding?
Claude Haiku 4.5 scores higher on coding benchmarks: 44.0 versus 32.9 in the Noometry coding category.
How many benchmarks do Claude 3.5 Haiku and Claude Haiku 4.5 share?
32 benchmarks have published results for both models. Claude 3.5 Haiku has 49 scored results on Noometry and Claude Haiku 4.5 has 53.