Model comparison
Llama 3.1-8B vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 23.0 on the Noometry Index. Llama 3.1-8B costs 25× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.
Last verified . 15 shared benchmarks.
Meta
StepFunSummary
- They share 15 benchmarks with published results for both. Llama 3.1-8B scores higher in 0 categories and Step 5 Preview in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Step 5 Preview leads 46.1 to 10.2.
- The biggest single-benchmark swing is SciCode: 13.2% for Llama 3.1-8B and 58.9% for Step 5 Preview.
- Llama 3.1-8B is cheaper at $0.05 / $0.08 per million input/output tokens, against $1 / $2.70 for Step 5 Preview.
- Step 5 Preview accepts more context: 1.02M tokens versus 128K.
- Llama 3.1-8B has downloadable open weights; the other is API-only.
Side by side
| Llama 3.1-8B | Step 5 Preview | |
|---|---|---|
| Provider | Meta | StepFun |
| Noometry Index | 23.0 | 47.9 |
| Released | 2024-07-23 | 2026-09-16 |
| Weights | Open | Proprietary |
| Context window | 128K | 1.02M |
| Max output | 4K | 66K |
| Input $ / M tokens | $0.05 | $1 |
| Output $ / M tokens | $0.08 | $2.70 |
| Results tracked | 43 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
Llama 3.1-8B: 20.2 (#340), Step 5 Preview: 51.8 (#36)
| Benchmark | Llama 3.1-8B | Step 5 Preview |
|---|---|---|
| SciCode | 13.2% | 58.9% |
| LMArena Coding | 1195 | 1480 |
| LMArena WebDev | — | 1564 |
| WeirdML | 1.7% | — |
| BigCodeBench Instruct | 32.8% | — |
| BigCodeBench Complete | 40.5% | — |
| HumanEval+ | 62.8% | — |
| MBPP+ | 55.6% | — |
Agentic & Tool Use Not comparable
Llama 3.1-8B: 22.5 (#131), Step 5 Preview: —
| Benchmark | Llama 3.1-8B | Step 5 Preview |
|---|---|---|
| Berkeley Function Calling Leaderboard | 25.8% | — |
| BALROG | 15.1% | — |
Reasoning Step 5 Preview leads
Llama 3.1-8B: 14.9 (#321), Step 5 Preview: 40.0 (#57)
| Benchmark | Llama 3.1-8B | Step 5 Preview |
|---|---|---|
| CritPt | 0% | 20.9% |
| LMArena Hard Prompts | 1175 | 1465 |
| Chess Puzzles | 0% | — |
| DTBench | 50.9% | — |
| LMCA | 5.4% | — |
| Epoch Capabilities Index | 116.57 | — |
| PIQA | 81.2% | — |
Math Step 5 Preview leads
Llama 3.1-8B: 10.2 (#317), Step 5 Preview: 46.1 (#72)
| Benchmark | Llama 3.1-8B | Step 5 Preview |
|---|---|---|
| LMArena Math | 1179 | 1470 |
| OTIS Mock AIME 2024-2025 | 1.7% | — |
| ProofBench | — | 42% |
| Omni-MATH | 13.7% | — |
| MATH Level 5 | 22.9% | — |
| GSM8K | 82.4% | — |
Knowledge Step 5 Preview leads
Llama 3.1-8B: 8.0 (#307), Step 5 Preview: 41.2 (#112)
| Benchmark | Llama 3.1-8B | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1144 | 1470 |
| GPQA Diamond | 27% | — |
| MMLU-Pro | 40.6% | — |
| GPQA (HELM) | 24.7% | — |
| BoolQ | 82.8% | — |
| MMLU | 56.1% | — |
Multimodal Not comparable
Llama 3.1-8B: —, Step 5 Preview: 41.0 (#41)
| Benchmark | Llama 3.1-8B | Step 5 Preview |
|---|---|---|
| LMArena Vision | — | 1267 |
Multilingual Step 5 Preview leads
Llama 3.1-8B: 34.0 (#249), Step 5 Preview: 53.8 (#54)
| Benchmark | Llama 3.1-8B | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1148 | 1432 |
| LMArena Chinese | 1151 | 1519 |
| LMArena Russian | 1158 | 1436 |
| LMArena Spanish | 1169 | 1439 |
| LMArena French | 1177 | — |
| LMArena German | 1144 | — |
| LMArena Japanese | 1061 | — |
| LMArena Korean | 1053 | — |
Instruction Following Step 5 Preview leads
Llama 3.1-8B: 58.9 (#258), Step 5 Preview: 76.0 (#50)
| Benchmark | Llama 3.1-8B | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1159 | 1444 |
| IFEval | 74.3% | — |
Long Context Step 5 Preview leads
Llama 3.1-8B: 35.8 (#238), Step 5 Preview: 44.5 (#64)
| Benchmark | Llama 3.1-8B | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1182 | 1455 |
Writing & Preference Step 5 Preview leads
Llama 3.1-8B: 29.7 (#290), Step 5 Preview: 62.8 (#71)
| Benchmark | Llama 3.1-8B | Step 5 Preview |
|---|---|---|
| LMArena Text | 1187 | 1442 |
| LMArena Creative Writing | 1154 | 1410 |
| LMArena Multi-Turn | 1172 | 1447 |
| EQ-Bench Creative Writing | 713 | — |
| WildBench | 68.7% | — |
Frequently asked questions
Is Llama 3.1-8B better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 23.0 on the Noometry Index. Llama 3.1-8B costs 25× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.
Which is cheaper, Llama 3.1-8B or Step 5 Preview?
Llama 3.1-8B is cheaper. It lists at $0.05 per million input tokens and $0.08 per million output tokens; Step 5 Preview lists at $1 and $2.70.
Is Llama 3.1-8B or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 20.2 in the Noometry coding category.
Which has the bigger context window?
Step 5 Preview does, with 1.02M tokens against 128K.
How many benchmarks do Llama 3.1-8B and Step 5 Preview share?
15 benchmarks have published results for both models. Llama 3.1-8B has 43 scored results on Noometry and Step 5 Preview has 18.