Model comparison
Codellama 34b Instruct vs DeepSeek-V2.5 (Sep 2024)
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 30.8 on the Noometry Index.
Last verified . 14 shared benchmarks.
Meta
DeepSeekSummary
- They share 14 benchmarks with published results for both. Codellama 34b Instruct scores higher in 0 categories and DeepSeek-V2.5 (Sep 2024) in 7 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V2.5 (Sep 2024) leads 49.8 to 28.2.
- The biggest single-benchmark swing is BigCodeBench Instruct: 29% for Codellama 34b Instruct and 48.6% for DeepSeek-V2.5 (Sep 2024).
Side by side
| Codellama 34b Instruct | DeepSeek-V2.5 (Sep 2024) | |
|---|---|---|
| Provider | Meta | DeepSeek |
| Noometry Index | 30.8 | 37.6 |
| Released | — | 2024-09-06 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 14 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V2.5 (Sep 2024) leads
Codellama 34b Instruct: 28.5 (#314), DeepSeek-V2.5 (Sep 2024): 31.7 (#281)
| Benchmark | Codellama 34b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| BigCodeBench Instruct | 29% | 48.6% |
| LMArena Coding | 1046 | 1309 |
| BigCodeBench Complete | 37.1% | 53.2% |
| HumanEval+ | 43.9% | 83.5% |
| MBPP+ | 56.3% | 74.1% |
| Aider Polyglot | — | 17.8% |
Reasoning DeepSeek-V2.5 (Sep 2024) leads
Codellama 34b Instruct: 19.6 (#255), DeepSeek-V2.5 (Sep 2024): 25.6 (#145)
| Benchmark | Codellama 34b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Hard Prompts | 1032 | 1289 |
Math DeepSeek-V2.5 (Sep 2024) leads
Codellama 34b Instruct: 31.0 (#230), DeepSeek-V2.5 (Sep 2024): 35.9 (#177)
| Benchmark | Codellama 34b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Math | 1056 | 1288 |
Knowledge Not comparable
Codellama 34b Instruct: —, DeepSeek-V2.5 (Sep 2024): 34.8 (#193)
| Benchmark | Codellama 34b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Expert | — | 1266 |
Multilingual DeepSeek-V2.5 (Sep 2024) leads
Codellama 34b Instruct: 25.8 (#284), DeepSeek-V2.5 (Sep 2024): 42.5 (#193)
| Benchmark | Codellama 34b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Non-English | 1011 | 1273 |
| LMArena Chinese | 976 | 1318 |
| LMArena French | — | 1289 |
| LMArena German | — | 1258 |
| LMArena Japanese | — | 1228 |
| LMArena Korean | — | 1209 |
| LMArena Russian | — | 1289 |
| LMArena Spanish | — | 1248 |
Instruction Following DeepSeek-V2.5 (Sep 2024) leads
Codellama 34b Instruct: 52.2 (#291), DeepSeek-V2.5 (Sep 2024): 67.5 (#194)
| Benchmark | Codellama 34b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Instruction Following | 1028 | 1280 |
Long Context DeepSeek-V2.5 (Sep 2024) leads
Codellama 34b Instruct: 30.9 (#284), DeepSeek-V2.5 (Sep 2024): 39.5 (#174)
| Benchmark | Codellama 34b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Longer Query | 1013 | 1301 |
Writing & Preference DeepSeek-V2.5 (Sep 2024) leads
Codellama 34b Instruct: 28.2 (#297), DeepSeek-V2.5 (Sep 2024): 49.8 (#187)
| Benchmark | Codellama 34b Instruct | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Text | 1066 | 1294 |
| LMArena Creative Writing | 1032 | 1285 |
| LMArena Multi-Turn | 1015 | 1297 |
Frequently asked questions
Is Codellama 34b Instruct better than DeepSeek-V2.5 (Sep 2024)?
DeepSeek-V2.5 (Sep 2024) is the stronger model overall, scoring 37.6 to 30.8 on the Noometry Index.
Is Codellama 34b Instruct or DeepSeek-V2.5 (Sep 2024) better for coding?
DeepSeek-V2.5 (Sep 2024) scores higher on coding benchmarks: 31.7 versus 28.5 in the Noometry coding category.
How many benchmarks do Codellama 34b Instruct and DeepSeek-V2.5 (Sep 2024) share?
14 benchmarks have published results for both models. Codellama 34b Instruct has 14 scored results on Noometry and DeepSeek-V2.5 (Sep 2024) has 22.