# Codellama 70b Instruct vs Step 3

> Step 3 is the stronger model overall, scoring 39.9 to 33.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/codellama-70b-instruct-vs-step-3
- Last updated: 2026-10-10
- Shared benchmarks: 4

## Summary

- They share 4 benchmarks with published results for both. Codellama 70b Instruct scores higher in 0 categories and Step 3 in 5 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Step 3 leads 46.5 to 24.9.

## Snapshot

| | Codellama 70b Instruct | Step 3 |
|---|---|---|
| Provider | Meta | StepFun |
| Noometry Index | 33.3 | 39.9 |
| Rank | 236 | 145 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Open | Open |

## Coding

- Codellama 70b Instruct: 37.0 (#189)
- Step 3: 39.5 (#144)

| Benchmark | Codellama 70b Instruct | Step 3 |
|---|---|---|
| BigCodeBench Instruct | 40.7% | — |
| LMArena Coding | — | 1367 |
| BigCodeBench Complete | 49.6% | — |
| HumanEval+ | 65.9% | — |

## Reasoning

- Codellama 70b Instruct: 19.2 (#241)
- Step 3: 27.6 (#102)

| Benchmark | Codellama 70b Instruct | Step 3 |
|---|---|---|
| LMArena Hard Prompts | 1052 | 1355 |
| Kagi LLM Benchmark | — | 62.3% |

## Math

- Codellama 70b Instruct: —
- Step 3: 34.9 (#148)

| Benchmark | Codellama 70b Instruct | Step 3 |
|---|---|---|
| LMArena Math | — | 1366 |

## Knowledge

- Codellama 70b Instruct: —
- Step 3: 36.2 (#161)

| Benchmark | Codellama 70b Instruct | Step 3 |
|---|---|---|
| LMArena Expert | — | 1333 |

## Multimodal

- Codellama 70b Instruct: —
- Step 3: 34.9 (#81)

| Benchmark | Codellama 70b Instruct | Step 3 |
|---|---|---|
| LMArena Vision | — | 1176 |

## Multilingual

- Codellama 70b Instruct: 24.9 (#283)
- Step 3: 46.5 (#154)

| Benchmark | Codellama 70b Instruct | Step 3 |
|---|---|---|
| LMArena Non-English | 992 | 1327 |
| LMArena Chinese | — | 1396 |
| LMArena German | — | 1371 |
| LMArena Korean | — | 1271 |
| LMArena Russian | — | 1331 |
| LMArena Spanish | — | 1372 |

## Instruction Following

- Codellama 70b Instruct: 52.1 (#288)
- Step 3: 70.6 (#159)

| Benchmark | Codellama 70b Instruct | Step 3 |
|---|---|---|
| LMArena Instruction Following | 1024 | 1332 |

## Long Context

- Codellama 70b Instruct: —
- Step 3: 40.3 (#152)

| Benchmark | Codellama 70b Instruct | Step 3 |
|---|---|---|
| LMArena Longer Query | — | 1326 |

## Writing & Preference

- Codellama 70b Instruct: 33.6 (#272)
- Step 3: 54.5 (#146)

| Benchmark | Codellama 70b Instruct | Step 3 |
|---|---|---|
| LMArena Text | 1057 | 1351 |
| LMArena Creative Writing | — | 1322 |
| LMArena Multi-Turn | — | 1340 |

## FAQ

### Is Codellama 70b Instruct better than Step 3?

Step 3 is the stronger model overall, scoring 39.9 to 33.3 on the Noometry Index.

### Is Codellama 70b Instruct or Step 3 better for coding?

Step 3 scores higher on coding benchmarks: 39.5 versus 37.0 in the Noometry coding category.

### How many benchmarks do Codellama 70b Instruct and Step 3 share?

4 benchmarks have published results for both models. Codellama 70b Instruct has 7 scored results on Noometry and Step 3 has 17.