Model comparison
Llama 3.1-70B vs Step 5 Preview
Step 5 Preview is the stronger model overall, scoring 47.9 to 29.6 on the Noometry Index. Llama 3.1-70B costs 3.6× less per token, which makes it the better buy when Step 5 Preview's lead doesn't matter for your workload.
Last verified . 13 shared benchmarks.
Meta
StepFunSummary
- They share 13 benchmarks with published results for both. Llama 3.1-70B 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 13.5.
- Llama 3.1-70B is cheaper at $0.40 / $0.40 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 3.1-70B has downloadable open weights; the other is API-only.
Side by side
| Llama 3.1-70B | Step 5 Preview | |
|---|---|---|
| Provider | Meta | StepFun |
| Noometry Index | 29.6 | 47.9 |
| Released | 2024-07-23 | 2026-09-16 |
| Weights | Open | Proprietary |
| Context window | 128K | 1.02M |
| Max output | 4K | 66K |
| Input $ / M tokens | $0.40 | $1 |
| Output $ / M tokens | $0.40 | $2.70 |
| Results tracked | 35 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Step 5 Preview leads
Llama 3.1-70B: 30.3 (#296), Step 5 Preview: 51.8 (#36)
| Benchmark | Llama 3.1-70B | Step 5 Preview |
|---|---|---|
| LMArena Coding | 1260 | 1480 |
| LMArena WebDev | — | 1564 |
| SciCode | — | 58.9% |
| WeirdML | 9% | — |
| BigCodeBench Instruct | 46.1% | — |
| BigCodeBench Complete | 54.8% | — |
Agentic & Tool Use Not comparable
Llama 3.1-70B: 25.1 (#112), Step 5 Preview: —
| Benchmark | Llama 3.1-70B | Step 5 Preview |
|---|---|---|
| TheAgentCompany | 6.9% | — |
| BALROG | 27.9% | — |
Reasoning Step 5 Preview leads
Llama 3.1-70B: 21.6 (#220), Step 5 Preview: 40.0 (#57)
| Benchmark | Llama 3.1-70B | Step 5 Preview |
|---|---|---|
| LMArena Hard Prompts | 1241 | 1465 |
| CritPt | — | 20.9% |
| DTBench | 60% | — |
| LMCA | 14.8% | — |
| Epoch Capabilities Index | 125.92 | — |
Math Step 5 Preview leads
Llama 3.1-70B: 13.5 (#304), Step 5 Preview: 46.1 (#72)
| Benchmark | Llama 3.1-70B | Step 5 Preview |
|---|---|---|
| LMArena Math | 1252 | 1470 |
| OTIS Mock AIME 2024-2025 | 3.6% | — |
| ProofBench | — | 42% |
| Omni-MATH | 21% | — |
| MATH Level 5 | 36.7% | — |
Knowledge Step 5 Preview leads
Llama 3.1-70B: 24.2 (#269), Step 5 Preview: 41.2 (#112)
| Benchmark | Llama 3.1-70B | Step 5 Preview |
|---|---|---|
| LMArena Expert | 1209 | 1470 |
| GPQA Diamond | 44.2% | — |
| MMLU-Pro | 65.3% | — |
| GPQA (HELM) | 42.6% | — |
| MMLU | 80.1% | — |
Multimodal Not comparable
Llama 3.1-70B: —, Step 5 Preview: 41.0 (#41)
| Benchmark | Llama 3.1-70B | Step 5 Preview |
|---|---|---|
| LMArena Vision | — | 1267 |
Multilingual Step 5 Preview leads
Llama 3.1-70B: 38.8 (#225), Step 5 Preview: 53.8 (#54)
| Benchmark | Llama 3.1-70B | Step 5 Preview |
|---|---|---|
| LMArena Non-English | 1219 | 1432 |
| LMArena Chinese | 1215 | 1519 |
| LMArena Russian | 1234 | 1436 |
| LMArena Spanish | 1253 | 1439 |
| LMArena French | 1261 | — |
| LMArena German | 1222 | — |
| LMArena Japanese | 1132 | — |
| LMArena Korean | 1140 | — |
Instruction Following Step 5 Preview leads
Llama 3.1-70B: 65.3 (#223), Step 5 Preview: 76.0 (#50)
| Benchmark | Llama 3.1-70B | Step 5 Preview |
|---|---|---|
| LMArena Instruction Following | 1231 | 1444 |
| IFEval | 82.1% | — |
Long Context Step 5 Preview leads
Llama 3.1-70B: 37.6 (#214), Step 5 Preview: 44.5 (#64)
| Benchmark | Llama 3.1-70B | Step 5 Preview |
|---|---|---|
| LMArena Longer Query | 1241 | 1455 |
Writing & Preference Step 5 Preview leads
Llama 3.1-70B: 35.4 (#267), Step 5 Preview: 62.8 (#71)
| Benchmark | Llama 3.1-70B | Step 5 Preview |
|---|---|---|
| LMArena Text | 1261 | 1442 |
| LMArena Creative Writing | 1232 | 1410 |
| LMArena Multi-Turn | 1256 | 1447 |
| EQ-Bench Creative Writing | 784 | — |
| WildBench | 75.8% | — |
Frequently asked questions
Is Llama 3.1-70B better than Step 5 Preview?
Step 5 Preview is the stronger model overall, scoring 47.9 to 29.6 on the Noometry Index. Llama 3.1-70B costs 3.6× 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 3.1-70B or Step 5 Preview?
Llama 3.1-70B is cheaper. It lists at $0.40 per million input tokens and $0.40 per million output tokens; Step 5 Preview lists at $1 and $2.70.
Is Llama 3.1-70B or Step 5 Preview better for coding?
Step 5 Preview scores higher on coding benchmarks: 51.8 versus 30.3 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 3.1-70B and Step 5 Preview share?
13 benchmarks have published results for both models. Llama 3.1-70B has 35 scored results on Noometry and Step 5 Preview has 18.