# Codellama 70b Instruct vs DeepSeek-V2.5 (Sep 2024)

> DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 33.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/codellama-70b-instruct-vs-deepseek-v2-5
- Last updated: 2026-10-11
- Shared benchmarks: 7

## Summary

- They share 7 benchmarks with published results for both. Codellama 70b Instruct scores higher in 1 category and DeepSeek-V2.5 (Sep 2024) in 4 categories; 5 gaps are clear of the uncertainty.
- The widest gap is in multilingual, where DeepSeek-V2.5 (Sep 2024) leads 42.5 to 24.8.
- The biggest single-benchmark swing is BigCodeBench Instruct: 40.7% for Codellama 70b Instruct and 48.6% for DeepSeek-V2.5 (Sep 2024).

## Snapshot

| | Codellama 70b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| Provider | Meta | DeepSeek |
| Noometry Index | 33.7 | 37.6 |
| Rank | 237 | 200 |
| Context | — | — |
| Input $/M | — | — |
| Output $/M | — | — |
| Weights | Open | Open |

## Coding

- Codellama 70b Instruct: 37.6 (#193)
- DeepSeek-V2.5 (Sep 2024): 31.7 (#281)

| Benchmark | Codellama 70b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| BigCodeBench Instruct | 40.7% | 48.6% |
| BigCodeBench Complete | 49.6% | 53.2% |
| HumanEval+ | 65.9% | 83.5% |
| Aider Polyglot | — | 17.8% |
| LMArena Coding | — | 1309 |
| MBPP+ | — | 74.1% |

## Reasoning

- Codellama 70b Instruct: 20.1 (#242)
- DeepSeek-V2.5 (Sep 2024): 25.6 (#145)

| Benchmark | Codellama 70b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Hard Prompts | 1052 | 1289 |

## Math

- Codellama 70b Instruct: —
- DeepSeek-V2.5 (Sep 2024): 35.9 (#177)

| Benchmark | Codellama 70b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Math | — | 1288 |

## Knowledge

- Codellama 70b Instruct: —
- DeepSeek-V2.5 (Sep 2024): 34.8 (#193)

| Benchmark | Codellama 70b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Expert | — | 1266 |

## Multilingual

- Codellama 70b Instruct: 24.8 (#288)
- DeepSeek-V2.5 (Sep 2024): 42.5 (#193)

| Benchmark | Codellama 70b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Non-English | 992 | 1273 |
| LMArena Chinese | — | 1318 |
| LMArena French | — | 1289 |
| LMArena German | — | 1258 |
| LMArena Japanese | — | 1228 |
| LMArena Korean | — | 1209 |
| LMArena Russian | — | 1289 |
| LMArena Spanish | — | 1248 |

## Instruction Following

- Codellama 70b Instruct: 51.9 (#293)
- DeepSeek-V2.5 (Sep 2024): 67.5 (#194)

| Benchmark | Codellama 70b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Instruction Following | 1024 | 1280 |

## Long Context

- Codellama 70b Instruct: —
- DeepSeek-V2.5 (Sep 2024): 39.5 (#174)

| Benchmark | Codellama 70b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Longer Query | — | 1301 |

## Writing & Preference

- Codellama 70b Instruct: 33.4 (#277)
- DeepSeek-V2.5 (Sep 2024): 49.8 (#187)

| Benchmark | Codellama 70b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Text | 1057 | 1294 |
| LMArena Creative Writing | — | 1285 |
| LMArena Multi-Turn | — | 1297 |

## FAQ

### Is Codellama 70b Instruct better than DeepSeek-V2.5 (Sep 2024)?

DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 33.7 on the Noometry Index.

### Is Codellama 70b Instruct or DeepSeek-V2.5 (Sep 2024) better for coding?

Codellama 70b Instruct scores higher on coding benchmarks: 37.6 versus 31.7 in the Noometry coding category.

### How many benchmarks do Codellama 70b Instruct and DeepSeek-V2.5 (Sep 2024) share?

7 benchmarks have published results for both models. Codellama 70b Instruct has 7 scored results on Noometry and DeepSeek-V2.5 (Sep 2024) has 22.