Model comparison
DeepSeek Coder 6.7B vs DeepSeek-V2.5 (Sep 2024)
DeepSeek-V2.5 (Sep 2024) has enough public results to be ranked (#200); DeepSeek Coder 6.7B does not yet, so treat this comparison as directional.
Last verified . 4 shared benchmarks.
DeepSeekSummary
- They share 4 benchmarks with published results for both. DeepSeek Coder 6.7B scores higher in 1 category and DeepSeek-V2.5 (Sep 2024) in 0 categories; one gap is clear of the uncertainty.
- The widest gap is in coding, where DeepSeek Coder 6.7B leads 35.8 to 31.7.
- The biggest single-benchmark swing is BigCodeBench Instruct: 35.5% for DeepSeek Coder 6.7B and 48.6% for DeepSeek-V2.5 (Sep 2024).
Side by side
| DeepSeek Coder 6.7B | DeepSeek-V2.5 (Sep 2024) | |
|---|---|---|
| Provider | DeepSeek | DeepSeek |
| Noometry Index | 37.6 | 37.6 |
| Released | 2023-11-02 | 2024-09-06 |
| Weights | Open | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 9 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek Coder 6.7B leads
DeepSeek Coder 6.7B: 35.8 (#219), DeepSeek-V2.5 (Sep 2024): 31.7 (#281)
| Benchmark | DeepSeek Coder 6.7B | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| BigCodeBench Instruct | 35.5% | 48.6% |
| BigCodeBench Complete | 43.8% | 53.2% |
| HumanEval+ | 71.3% | 83.5% |
| MBPP+ | 65.6% | 74.1% |
| Aider Polyglot | — | 17.8% |
| LMArena Coding | — | 1309 |
Reasoning Not comparable
DeepSeek Coder 6.7B: —, DeepSeek-V2.5 (Sep 2024): 25.6 (#145)
| Benchmark | DeepSeek Coder 6.7B | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Hard Prompts | — | 1289 |
| Epoch Capabilities Index | 89.62 | — |
| WinoGrande | 57.6% | — |
Math Not comparable
DeepSeek Coder 6.7B: —, DeepSeek-V2.5 (Sep 2024): 35.9 (#177)
| Benchmark | DeepSeek Coder 6.7B | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Math | — | 1288 |
| GSM8K | 21.3% | — |
Knowledge Not comparable
DeepSeek Coder 6.7B: —, DeepSeek-V2.5 (Sep 2024): 34.8 (#193)
| Benchmark | DeepSeek Coder 6.7B | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Expert | — | 1266 |
| ARC (AI2) Challenge | 36.4% | — |
| MMLU | 36.4% | — |
Multilingual Not comparable
DeepSeek Coder 6.7B: —, DeepSeek-V2.5 (Sep 2024): 42.5 (#193)
| Benchmark | DeepSeek Coder 6.7B | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Non-English | — | 1273 |
| LMArena Chinese | — | 1318 |
| LMArena French | — | 1289 |
| LMArena German | — | 1258 |
| LMArena Japanese | — | 1228 |
| LMArena Korean | — | 1209 |
| LMArena Russian | — | 1289 |
| LMArena Spanish | — | 1248 |
Instruction Following Not comparable
DeepSeek Coder 6.7B: —, DeepSeek-V2.5 (Sep 2024): 67.5 (#194)
| Benchmark | DeepSeek Coder 6.7B | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Instruction Following | — | 1280 |
Long Context Not comparable
DeepSeek Coder 6.7B: —, DeepSeek-V2.5 (Sep 2024): 39.5 (#174)
| Benchmark | DeepSeek Coder 6.7B | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Longer Query | — | 1301 |
Writing & Preference Not comparable
DeepSeek Coder 6.7B: —, DeepSeek-V2.5 (Sep 2024): 49.8 (#187)
| Benchmark | DeepSeek Coder 6.7B | DeepSeek-V2.5 (Sep 2024) |
|---|---|---|
| LMArena Text | — | 1294 |
| LMArena Creative Writing | — | 1285 |
| LMArena Multi-Turn | — | 1297 |
Frequently asked questions
Is DeepSeek Coder 6.7B better than DeepSeek-V2.5 (Sep 2024)?
DeepSeek-V2.5 (Sep 2024) has enough public results to be ranked (#200); DeepSeek Coder 6.7B does not yet, so treat this comparison as directional.
Is DeepSeek Coder 6.7B or DeepSeek-V2.5 (Sep 2024) better for coding?
DeepSeek Coder 6.7B scores higher on coding benchmarks: 35.8 versus 31.7 in the Noometry coding category.
How many benchmarks do DeepSeek Coder 6.7B and DeepSeek-V2.5 (Sep 2024) share?
4 benchmarks have published results for both models. DeepSeek Coder 6.7B has 9 scored results on Noometry and DeepSeek-V2.5 (Sep 2024) has 22.