Model comparison
Claude 3 Haiku vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 25.9 on the Noometry Index.
Last verified . 14 shared benchmarks.
StepFunSummary
- They share 14 benchmarks with published results for both. Claude 3 Haiku scores higher in 0 categories and Step 5 Preview in 9 categories; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where Step 5 Preview leads 46.1 to 9.8.
Side by side
| Claude 3 Haiku | Step 5 Preview | |
|---|---|---|
| Provider | Anthropic | StepFun |
| Noometry Index | 25.9 | 47.9 |
| Released | 2024-03-07 | 2026-09-16 |
| Weights | Proprietary | Proprietary |
| Context window | — | 1.02M |
| Max output | — | 66K |
| Input $ / M tokens | — | $1 |
| Output $ / M tokens | — | $2.70 |
| Results tracked | 37 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
Claude 3 Haiku: 26.4 (#325), Step 5 Preview: 51.8 (#36)
| Benchmark | Claude 3 Haiku | Step 5 Preview |
|---|---|---|
| LMArena Coding | 1199 | 1480 |
| LMArena WebDev | — | 1564 |
| SciCode | — | 58.9% |
| WeirdML | 9.8% | — |
| BigCodeBench Instruct | 39.4% | — |
| BigCodeBench Complete | 50.1% | — |
| CadEval | 12% | — |
| HumanEval+ | 68.9% | — |
| MBPP+ | 68.8% | — |
Reasoning Step 5 Preview leads
Claude 3 Haiku: 16.3 (#307), Step 5 Preview: 40.0 (#57)
| Benchmark | Claude 3 Haiku | Step 5 Preview |
|---|---|---|
| LMArena Hard Prompts | 1174 | 1465 |
| Kagi LLM Benchmark | 34.2% | — |
| CritPt | — | 20.9% |
| DTBench | 50.1% | — |
| LMCA | 8.8% | — |
| Epoch Capabilities Index | 118.35 | — |
| ForecastBench | 53.2 | — |
| WinoGrande | 74.2% | — |
Math Step 5 Preview leads
Claude 3 Haiku: 9.8 (#319), Step 5 Preview: 46.1 (#72)
| Benchmark | Claude 3 Haiku | Step 5 Preview |
|---|---|---|
| LMArena Math | 1188 | 1470 |
| OTIS Mock AIME 2024-2025 | 1.8% | — |
| ProofBench | — | 42% |
| MATH Level 5 | 14.9% | — |
Knowledge Step 5 Preview leads
Claude 3 Haiku: 17.3 (#285), Step 5 Preview: 41.2 (#112)
| Benchmark | Claude 3 Haiku | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1148 | 1470 |
| GPQA Diamond | 36.3% | — |
| Confabulations | 34.2% | — |
| MMLU | 73.8% | — |
Multimodal Step 5 Preview leads
Claude 3 Haiku: 23.6 (#128), Step 5 Preview: 41.0 (#41)
| Benchmark | Claude 3 Haiku | Step 5 Preview |
|---|---|---|
| LMArena Vision | 950 | 1267 |
| ScienceQA | 72% | — |
Multilingual Step 5 Preview leads
Claude 3 Haiku: 36.0 (#243), Step 5 Preview: 53.8 (#54)
| Benchmark | Claude 3 Haiku | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1178 | 1432 |
| LMArena Chinese | 1155 | 1519 |
| LMArena Russian | 1204 | 1436 |
| LMArena Spanish | 1166 | 1439 |
| LMArena French | 1195 | — |
| LMArena German | 1174 | — |
| LMArena Japanese | 1102 | — |
| LMArena Korean | 1109 | — |
Instruction Following Step 5 Preview leads
Claude 3 Haiku: 61.3 (#247), Step 5 Preview: 76.0 (#50)
| Benchmark | Claude 3 Haiku | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1173 | 1444 |
Long Context Step 5 Preview leads
Claude 3 Haiku: 36.1 (#237), Step 5 Preview: 44.5 (#64)
| Benchmark | Claude 3 Haiku | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1190 | 1455 |
Writing & Preference Step 5 Preview leads
Claude 3 Haiku: 29.7 (#291), Step 5 Preview: 62.8 (#71)
| Benchmark | Claude 3 Haiku | Step 5 Preview |
|---|---|---|
| LMArena Text | 1195 | 1442 |
| LMArena Creative Writing | 1157 | 1410 |
| LMArena Multi-Turn | 1190 | 1447 |
| EQ-Bench Creative Writing | 717 | — |
Frequently asked questions
Is Claude 3 Haiku better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 25.9 on the Noometry Index.
Is Claude 3 Haiku or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 26.4 in the Noometry coding category.
How many benchmarks do Claude 3 Haiku and Step 5 Preview share?
14 benchmarks have published results for both models. Claude 3 Haiku has 37 scored results on Noometry and Step 5 Preview has 18.