Model comparison
DeepSeek-V2 (MoE-236B, May 2024) vs GPT-4 Turbo
GPT-4 Turbo has enough public results to be ranked (#281); DeepSeek-V2 (MoE-236B, May 2024) 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-V2 (MoE-236B, May 2024) scores higher in 1 category and GPT-4 Turbo in 0 categories; one gap is clear of the uncertainty.
- The widest gap is in coding, where DeepSeek-V2 (MoE-236B, May 2024) leads 39.8 to 33.4.
- DeepSeek-V2 (MoE-236B, May 2024) has downloadable open weights; the other is API-only.
Side by side
| DeepSeek-V2 (MoE-236B, May 2024) | GPT-4 Turbo | |
|---|---|---|
| Provider | DeepSeek | OpenAI |
| Noometry Index | 39.7 | 30.3 |
| Released | 2024-05-07 | 2023-11-06 |
| Weights | Open | Proprietary |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $10 |
| Output $ / M tokens | — | $30 |
| Results tracked | 10 | 36 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V2 (MoE-236B, May 2024) leads
DeepSeek-V2 (MoE-236B, May 2024): 39.8 (#141), GPT-4 Turbo: 33.4 (#247)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | GPT-4 Turbo |
|---|---|---|
| BigCodeBench Instruct | 48.9% | 48.2% |
| BigCodeBench Complete | 59.4% | 58.2% |
| WeirdML | — | 18% |
| LMArena Coding | — | 1268 |
| HumanEval+ | — | 86.6% |
| MBPP+ | — | 73.3% |
Agentic & Tool Use Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, GPT-4 Turbo: —
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | GPT-4 Turbo |
|---|---|---|
| METR Time Horizons | — | 36.7% |
Reasoning Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, GPT-4 Turbo: 14.9 (#311)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | GPT-4 Turbo |
|---|---|---|
| Epoch Capabilities Index | 124.78 | 127.25 |
| SimpleBench | — | 25.1% |
| Chess Puzzles | — | 6% |
| LMArena Hard Prompts | — | 1251 |
| DTBench | — | 61.6% |
| LMCA | — | 9.8% |
| BIG-Bench Hard | 78.8% | — |
| ForecastBench | — | 59.4 |
| HellaSwag | 87.1% | — |
| PIQA | 83.9% | — |
| WinoGrande | 86.3% | — |
Math Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, GPT-4 Turbo: 8.6 (#316)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | GPT-4 Turbo |
|---|---|---|
| FrontierMath (Tiers 1-3) | — | 0.7% |
| OTIS Mock AIME 2024-2025 | — | 6.7% |
| LMArena Math | — | 1272 |
| MATH Level 5 | — | 46.7% |
Knowledge Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, GPT-4 Turbo: 24.3 (#263)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | GPT-4 Turbo |
|---|---|---|
| MMLU | 78.4% | 81.3% |
| GPQA Diamond | — | 46.6% |
| Confabulations | — | 28.4% |
| LMArena Expert | — | 1223 |
| ARC (AI2) Challenge | 92.2% | — |
| TriviaQA | 80% | — |
Multimodal Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, GPT-4 Turbo: 30.0 (#104)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | GPT-4 Turbo |
|---|---|---|
| LMArena Vision | — | 1090 |
Multilingual Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, GPT-4 Turbo: 40.7 (#211)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | GPT-4 Turbo |
|---|---|---|
| LMArena Non-English | — | 1245 |
| LMArena Chinese | — | 1242 |
| LMArena French | — | 1275 |
| LMArena German | — | 1259 |
| LMArena Japanese | — | 1194 |
| LMArena Korean | — | 1187 |
| LMArena Russian | — | 1259 |
| LMArena Spanish | — | 1260 |
Instruction Following Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, GPT-4 Turbo: 66.0 (#211)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | GPT-4 Turbo |
|---|---|---|
| LMArena Instruction Following | — | 1250 |
Long Context Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, GPT-4 Turbo: 38.0 (#200)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | GPT-4 Turbo |
|---|---|---|
| LMArena Longer Query | — | 1254 |
Writing & Preference Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, GPT-4 Turbo: 47.9 (#201)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | GPT-4 Turbo |
|---|---|---|
| LMArena Text | — | 1272 |
| LMArena Creative Writing | — | 1269 |
| LMArena Multi-Turn | — | 1267 |
Frequently asked questions
Is DeepSeek-V2 (MoE-236B, May 2024) better than GPT-4 Turbo?
GPT-4 Turbo has enough public results to be ranked (#281); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Is DeepSeek-V2 (MoE-236B, May 2024) or GPT-4 Turbo better for coding?
DeepSeek-V2 (MoE-236B, May 2024) scores higher on coding benchmarks: 39.8 versus 33.4 in the Noometry coding category.
How many benchmarks do DeepSeek-V2 (MoE-236B, May 2024) and GPT-4 Turbo share?
4 benchmarks have published results for both models. DeepSeek-V2 (MoE-236B, May 2024) has 10 scored results on Noometry and GPT-4 Turbo has 36.