Model comparison
DeepSeek Coder 6.7B vs GPT-4 Turbo
GPT-4 Turbo has enough public results to be ranked (#292); DeepSeek Coder 6.7B does not yet, so treat this comparison as directional.
Last verified . 6 shared benchmarks.
DeepSeekSummary
- They share 6 benchmarks with published results for both. DeepSeek Coder 6.7B scores higher in 1 category and GPT-4 Turbo in 0 categories; one gap is clear of the uncertainty.
- The biggest single-benchmark swing is BigCodeBench Complete: 43.8% for DeepSeek Coder 6.7B and 58.2% for GPT-4 Turbo.
- DeepSeek Coder 6.7B has downloadable open weights; the other is API-only.
Side by side
| DeepSeek Coder 6.7B | GPT-4 Turbo | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 37.6 | 30.5 |
| Released | 2023-11-02 | 2023-11-06 |
| Weights | Open | Proprietary |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $10 |
| Output $ / M tokens | — | $30 |
| Results tracked | 9 | 36 |
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), GPT-4 Turbo: 33.8 (#249)
| Benchmark | DeepSeek Coder 6.7B | GPT-4 Turbo |
|---|---|---|
| BigCodeBench Instruct | 35.5% | 48.2% |
| BigCodeBench Complete | 43.8% | 58.2% |
| HumanEval+ | 71.3% | 86.6% |
| MBPP+ | 65.6% | 73.3% |
| WeirdML | — | 18% |
| LMArena Coding | — | 1268 |
Agentic & Tool Use Not comparable
DeepSeek Coder 6.7B: —, GPT-4 Turbo: —
| Benchmark | DeepSeek Coder 6.7B | GPT-4 Turbo |
|---|---|---|
| METR Time Horizons | — | 36.7% |
Reasoning Not comparable
DeepSeek Coder 6.7B: —, GPT-4 Turbo: 15.3 (#317)
| Benchmark | DeepSeek Coder 6.7B | GPT-4 Turbo |
|---|---|---|
| Epoch Capabilities Index | 89.62 | 127.25 |
| SimpleBench | — | 25.1% |
| Chess Puzzles | — | 6% |
| LMArena Hard Prompts | — | 1251 |
| DTBench | — | 61.6% |
| LMCA | — | 9.8% |
| ForecastBench | — | 59.4 |
| WinoGrande | 57.6% | — |
Math Not comparable
DeepSeek Coder 6.7B: —, GPT-4 Turbo: 9.0 (#322)
| Benchmark | DeepSeek Coder 6.7B | GPT-4 Turbo |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 0.7% |
| OTIS Mock AIME 2024-2025 | — | 6.7% |
| LMArena Math | — | 1272 |
| MATH Level 5 | — | 46.7% |
| GSM8K | 21.3% | — |
Knowledge Not comparable
DeepSeek Coder 6.7B: —, GPT-4 Turbo: 24.3 (#268)
| Benchmark | DeepSeek Coder 6.7B | GPT-4 Turbo |
|---|---|---|
| MMLU | 36.4% | 81.3% |
| GPQA Diamond | — | 46.6% |
| Confabulations | — | 28.4% |
| LMArena Expert | — | 1223 |
| ARC (AI2) Challenge | 36.4% | — |
Multimodal Not comparable
DeepSeek Coder 6.7B: —, GPT-4 Turbo: 30.6 (#110)
| Benchmark | DeepSeek Coder 6.7B | GPT-4 Turbo |
|---|---|---|
| LMArena Vision | — | 1090 |
Multilingual Not comparable
DeepSeek Coder 6.7B: —, GPT-4 Turbo: 40.5 (#216)
| Benchmark | DeepSeek Coder 6.7B | GPT-4 Turbo |
|---|---|---|
| LMArena Non-English | — | 1245 |
| LMArena Chinese | — | 1242 |
| LMArena French | — | 1276 |
| LMArena German | — | 1259 |
| LMArena Japanese | — | 1194 |
| LMArena Korean | — | 1187 |
| LMArena Russian | — | 1259 |
| LMArena Spanish | — | 1260 |
Instruction Following Not comparable
DeepSeek Coder 6.7B: —, GPT-4 Turbo: 65.8 (#216)
| Benchmark | DeepSeek Coder 6.7B | GPT-4 Turbo |
|---|---|---|
| LMArena Instruction Following | — | 1249 |
Long Context Not comparable
DeepSeek Coder 6.7B: —, GPT-4 Turbo: 38.0 (#206)
| Benchmark | DeepSeek Coder 6.7B | GPT-4 Turbo |
|---|---|---|
| LMArena Longer Query | — | 1254 |
Writing & Preference Not comparable
DeepSeek Coder 6.7B: —, GPT-4 Turbo: 47.7 (#206)
| Benchmark | DeepSeek Coder 6.7B | GPT-4 Turbo |
|---|---|---|
| LMArena Text | — | 1272 |
| LMArena Creative Writing | — | 1269 |
| LMArena Multi-Turn | — | 1267 |
Frequently asked questions
Is DeepSeek Coder 6.7B better than GPT-4 Turbo?
GPT-4 Turbo has enough public results to be ranked (#292); DeepSeek Coder 6.7B does not yet, so treat this comparison as directional.
Is DeepSeek Coder 6.7B or GPT-4 Turbo better for coding?
DeepSeek Coder 6.7B scores higher on coding benchmarks: 35.8 versus 33.8 in the Noometry coding category.
How many benchmarks do DeepSeek Coder 6.7B and GPT-4 Turbo share?
6 benchmarks have published results for both models. DeepSeek Coder 6.7B has 9 scored results on Noometry and GPT-4 Turbo has 36.