# Codellama 70b Instruct vs gpt-oss-20b

> Codellama 70b Instruct is the stronger model overall, scoring 33.3 to 32.1 on the Noometry Index.

- Canonical page: https://noometry.com/compare/codellama-70b-instruct-vs-gpt-oss-20b
- Last updated: 2026-10-10
- Shared benchmarks: 4

## Summary

- They share 4 benchmarks with published results for both. Codellama 70b Instruct scores higher in 1 category and gpt-oss-20b in 4 categories; 3 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where gpt-oss-20b leads 42.4 to 24.9.

## Snapshot

| | Codellama 70b Instruct | gpt-oss-20b |
|---|---|---|
| Provider | Meta | OpenAI |
| Noometry Index | 33.3 | 32.1 |
| Rank | 236 | 249 |
| Context | — | 131K |
| Input $/M | — | $0.018 |
| Output $/M | — | $0.09 |
| Weights | Open | Open |

## Coding

- Codellama 70b Instruct: 37.0 (#189)
- gpt-oss-20b: 37.1 (#185)

| Benchmark | Codellama 70b Instruct | gpt-oss-20b |
|---|---|---|
| SciCode | — | 34.4% |
| WeirdML | — | 40.9% |
| BigCodeBench Instruct | 40.7% | — |
| LMArena Coding | — | 1307 |
| BigCodeBench Complete | 49.6% | — |
| ALE-Bench | — | 566.05 |
| HumanEval+ | 65.9% | — |

## Agentic & Tool Use

- Codellama 70b Instruct: —
- gpt-oss-20b: 9.2 (#146)

| Benchmark | Codellama 70b Instruct | gpt-oss-20b |
|---|---|---|
| Terminal-Bench | — | 3.4% |

## Reasoning

- Codellama 70b Instruct: 19.2 (#241)
- gpt-oss-20b: 18.8 (#251)

| Benchmark | Codellama 70b Instruct | gpt-oss-20b |
|---|---|---|
| LMArena Hard Prompts | 1052 | 1274 |
| Kagi LLM Benchmark | — | 53.2% |
| CritPt | — | 1.4% |
| Chess Puzzles | — | 4% |
| DTBench | — | 68% |
| LMCA | — | 14.5% |
| Epoch Capabilities Index | — | 137.81 |

## Math

- Codellama 70b Instruct: —
- gpt-oss-20b: 37.6 (#93)

| Benchmark | Codellama 70b Instruct | gpt-oss-20b |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 65.3% |
| Omni-MATH | — | 56.5% |
| LMArena Math | — | 1317 |

## Knowledge

- Codellama 70b Instruct: —
- gpt-oss-20b: 34.6 (#186)

| Benchmark | Codellama 70b Instruct | gpt-oss-20b |
|---|---|---|
| GPQA Diamond | — | 60.8% |
| MMLU-Pro | — | 74% |
| GPQA (HELM) | — | 59.4% |
| LMArena Expert | — | 1258 |

## Multilingual

- Codellama 70b Instruct: 24.9 (#283)
- gpt-oss-20b: 42.4 (#192)

| Benchmark | Codellama 70b Instruct | gpt-oss-20b |
|---|---|---|
| LMArena Non-English | 992 | 1269 |
| LMArena Chinese | — | 1311 |
| LMArena German | — | 1249 |
| LMArena Japanese | — | 1245 |
| LMArena Korean | — | 1239 |
| LMArena Russian | — | 1279 |
| LMArena Spanish | — | 1267 |

## Instruction Following

- Codellama 70b Instruct: 52.1 (#288)
- gpt-oss-20b: 62.0 (#235)

| Benchmark | Codellama 70b Instruct | gpt-oss-20b |
|---|---|---|
| LMArena Instruction Following | 1024 | 1236 |
| IFEval | — | 73.2% |

## Long Context

- Codellama 70b Instruct: —
- gpt-oss-20b: 37.9 (#203)

| Benchmark | Codellama 70b Instruct | gpt-oss-20b |
|---|---|---|
| LMArena Longer Query | — | 1251 |

## Writing & Preference

- Codellama 70b Instruct: 33.6 (#272)
- gpt-oss-20b: 35.8 (#260)

| Benchmark | Codellama 70b Instruct | gpt-oss-20b |
|---|---|---|
| LMArena Text | 1057 | 1288 |
| LMArena Creative Writing | — | 1201 |
| EQ-Bench Creative Writing | — | 666 |
| WildBench | — | 73.7% |
| LMArena Multi-Turn | — | 1268 |

## FAQ

### Is Codellama 70b Instruct better than gpt-oss-20b?

Codellama 70b Instruct is the stronger model overall, scoring 33.3 to 32.1 on the Noometry Index.

### Is Codellama 70b Instruct or gpt-oss-20b better for coding?

They score almost the same on coding (37.0 vs 37.1); test both on your own repository before choosing.

### How many benchmarks do Codellama 70b Instruct and gpt-oss-20b share?

4 benchmarks have published results for both models. Codellama 70b Instruct has 7 scored results on Noometry and gpt-oss-20b has 34.