# Codellama 70b Instruct vs Qwen3 32B

> Qwen3 32B is the stronger model overall, scoring 38.8 to 33.3 on the Noometry Index.

- Canonical page: https://noometry.com/compare/codellama-70b-instruct-vs-qwen3-32b
- 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 Qwen3 32B in 5 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where Qwen3 32B leads 45.8 to 24.9.

## Snapshot

| | Codellama 70b Instruct | Qwen3 32B |
|---|---|---|
| Provider | Meta | Alibaba (Qwen) |
| Noometry Index | 33.3 | 38.8 |
| Rank | 236 | 166 |
| Context | — | 131K |
| Input $/M | — | $0.70 |
| Output $/M | — | $2.80 |
| Weights | Open | Open |

## Coding

- Codellama 70b Instruct: 37.0 (#189)
- Qwen3 32B: 37.3 (#184)

| Benchmark | Codellama 70b Instruct | Qwen3 32B |
|---|---|---|
| Aider Polyglot | — | 40% |
| SciCode | — | 35.4% |
| BigCodeBench Instruct | 40.7% | — |
| LMArena Coding | — | 1358 |
| BigCodeBench Complete | 49.6% | — |
| HumanEval+ | 65.9% | — |

## Agentic & Tool Use

- Codellama 70b Instruct: —
- Qwen3 32B: 32.3 (#58)

| Benchmark | Codellama 70b Instruct | Qwen3 32B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 48.7% |

## Reasoning

- Codellama 70b Instruct: 19.2 (#241)
- Qwen3 32B: 19.6 (#233)

| Benchmark | Codellama 70b Instruct | Qwen3 32B |
|---|---|---|
| LMArena Hard Prompts | 1052 | 1335 |
| Kagi LLM Benchmark | — | 54.9% |
| CritPt | — | 0.3% |
| Chess Puzzles | — | 5% |
| DTBench | — | 67.5% |
| LMCA | — | 17.3% |
| Epoch Capabilities Index | — | 138.51 |

## Math

- Codellama 70b Instruct: —
- Qwen3 32B: 37.9 (#89)

| Benchmark | Codellama 70b Instruct | Qwen3 32B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 66.9% |
| LMArena Math | — | 1399 |

## Knowledge

- Codellama 70b Instruct: —
- Qwen3 32B: 40.0 (#112)

| Benchmark | Codellama 70b Instruct | Qwen3 32B |
|---|---|---|
| GPQA Diamond | — | 65.7% |
| Vectara Hallucination Rate | — | 5.9% |
| LMArena Expert | — | 1362 |

## Multilingual

- Codellama 70b Instruct: 24.9 (#283)
- Qwen3 32B: 45.8 (#162)

| Benchmark | Codellama 70b Instruct | Qwen3 32B |
|---|---|---|
| LMArena Non-English | 992 | 1317 |
| LMArena Chinese | — | 1357 |
| LMArena German | — | 1341 |
| LMArena Russian | — | 1311 |

## Instruction Following

- Codellama 70b Instruct: 52.1 (#288)
- Qwen3 32B: 69.1 (#174)

| Benchmark | Codellama 70b Instruct | Qwen3 32B |
|---|---|---|
| LMArena Instruction Following | 1024 | 1305 |

## Long Context

- Codellama 70b Instruct: —
- Qwen3 32B: 43.7 (#83)

| Benchmark | Codellama 70b Instruct | Qwen3 32B |
|---|---|---|
| Fiction.LiveBench | — | 74.2% |
| LMArena Longer Query | — | 1327 |

## Writing & Preference

- Codellama 70b Instruct: 33.6 (#272)
- Qwen3 32B: 53.1 (#158)

| Benchmark | Codellama 70b Instruct | Qwen3 32B |
|---|---|---|
| LMArena Text | 1057 | 1340 |
| LMArena Creative Writing | — | 1297 |
| LMArena Multi-Turn | — | 1331 |

## FAQ

### Is Codellama 70b Instruct better than Qwen3 32B?

Qwen3 32B is the stronger model overall, scoring 38.8 to 33.3 on the Noometry Index.

### Is Codellama 70b Instruct or Qwen3 32B better for coding?

They score almost the same on coding (37.0 vs 37.3); test both on your own repository before choosing.

### How many benchmarks do Codellama 70b Instruct and Qwen3 32B share?

4 benchmarks have published results for both models. Codellama 70b Instruct has 7 scored results on Noometry and Qwen3 32B has 26.