Model comparison
Claude 3 Opus vs DeepSeek-V2 (MoE-236B, May 2024)
Claude 3 Opus has enough public results to be ranked (#310); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Last verified . 5 shared benchmarks.
DeepSeekSummary
- They share 5 benchmarks with published results for both. Claude 3 Opus scores higher in 0 categories and DeepSeek-V2 (MoE-236B, May 2024) in 1 category; one gap is clear of the uncertainty.
- The widest gap is in coding, where DeepSeek-V2 (MoE-236B, May 2024) leads 40.4 to 32.9.
- DeepSeek-V2 (MoE-236B, May 2024) has downloadable open weights; the other is API-only.
Side by side
| Claude 3 Opus | DeepSeek-V2 (MoE-236B, May 2024) | |
|---|---|---|
| Provider | Anthropic | DeepSeek |
| Noometry Index | 29.5 | 40.3 |
| Released | 2024-02-29 | 2024-05-07 |
| Weights | Proprietary | Open |
| Context window | — | — |
| Max output | — | — |
| Input $ / M tokens | — | — |
| Output $ / M tokens | — | — |
| Results tracked | 46 | 10 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V2 (MoE-236B, May 2024) leads
Claude 3 Opus: 32.9 (#267), DeepSeek-V2 (MoE-236B, May 2024): 40.4 (#139)
| Benchmark | Claude 3 Opus | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| BigCodeBench Instruct | 45.5% | 48.9% |
| BigCodeBench Complete | 57.4% | 59.4% |
| WeirdML | 19.2% | — |
| LiveBench Coding | 38.6% | — |
| LMArena Coding | 1264 | — |
| HumanEval+ | 77.4% | — |
| MBPP+ | 73.3% | — |
Agentic & Tool Use Not comparable
Claude 3 Opus: 24.6 (#116), DeepSeek-V2 (MoE-236B, May 2024): —
| Benchmark | Claude 3 Opus | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| Cybench | 10% | — |
| METR Time Horizons | 29.5% | — |
Reasoning Not comparable
Claude 3 Opus: 14.6 (#324), DeepSeek-V2 (MoE-236B, May 2024): —
| Benchmark | Claude 3 Opus | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| Epoch Capabilities Index | 126.91 | 124.77 |
| WinoGrande | 88.5% | 86.3% |
| SimpleBench | 23.5% | — |
| Chess Puzzles | 5% | — |
| EnigmaEval | 0.8% | — |
| LiveBench Reasoning | 40.6% | — |
| LMArena Hard Prompts | 1245 | — |
| DTBench | 61.6% | — |
| LiveBench Data Analysis | 57.9% | — |
| LMCA | 17% | — |
| BIG-Bench Hard | — | 78.8% |
| ForecastBench | 58.4 | — |
| HellaSwag | — | 87.1% |
| LiveBench | 49.2% | — |
| PIQA | — | 83.9% |
Math Not comparable
Claude 3 Opus: 14.8 (#299), DeepSeek-V2 (MoE-236B, May 2024): —
| Benchmark | Claude 3 Opus | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 4.7% | — |
| LiveBench Math | 43.6% | — |
| LMArena Math | 1273 | — |
| MATH Level 5 | 37.5% | — |
Knowledge Not comparable
Claude 3 Opus: 24.5 (#267), DeepSeek-V2 (MoE-236B, May 2024): —
| Benchmark | Claude 3 Opus | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| MMLU | 84.6% | 78.4% |
| GPQA Diamond | 47.2% | — |
| SimpleQA Verified | 12.6% | — |
| Confabulations | 22.7% | — |
| LMArena Expert | 1223 | — |
| ARC (AI2) Challenge | — | 92.2% |
| TriviaQA | — | 80% |
Multimodal Not comparable
Claude 3 Opus: 27.1 (#116), DeepSeek-V2 (MoE-236B, May 2024): —
| Benchmark | Claude 3 Opus | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| LMArena Vision | 1023 | — |
Multilingual Not comparable
Claude 3 Opus: 41.4 (#207), DeepSeek-V2 (MoE-236B, May 2024): —
| Benchmark | Claude 3 Opus | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| LMArena Non-English | 1258 | — |
| LMArena Chinese | 1248 | — |
| LMArena French | 1275 | — |
| LMArena German | 1258 | — |
| LMArena Japanese | 1204 | — |
| LMArena Korean | 1187 | — |
| LMArena Russian | 1280 | — |
| LMArena Spanish | 1246 | — |
Instruction Following Not comparable
Claude 3 Opus: 64.1 (#228), DeepSeek-V2 (MoE-236B, May 2024): —
| Benchmark | Claude 3 Opus | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| LiveBench Instruction Following | 63.9% | — |
| LMArena Instruction Following | 1248 | — |
Long Context Not comparable
Claude 3 Opus: 38.2 (#202), DeepSeek-V2 (MoE-236B, May 2024): —
| Benchmark | Claude 3 Opus | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| LMArena Longer Query | 1259 | — |
Writing & Preference Not comparable
Claude 3 Opus: 47.2 (#213), DeepSeek-V2 (MoE-236B, May 2024): —
| Benchmark | Claude 3 Opus | DeepSeek-V2 (MoE-236B, May 2024) |
|---|---|---|
| LMArena Text | 1262 | — |
| LMArena Creative Writing | 1235 | — |
| LMArena Multi-Turn | 1275 | — |
| LiveBench Language | 50.4% | — |
Frequently asked questions
Is Claude 3 Opus better than DeepSeek-V2 (MoE-236B, May 2024)?
Claude 3 Opus has enough public results to be ranked (#310); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Is Claude 3 Opus or DeepSeek-V2 (MoE-236B, May 2024) better for coding?
DeepSeek-V2 (MoE-236B, May 2024) scores higher on coding benchmarks: 40.4 versus 32.9 in the Noometry coding category.
How many benchmarks do Claude 3 Opus and DeepSeek-V2 (MoE-236B, May 2024) share?
5 benchmarks have published results for both models. Claude 3 Opus has 46 scored results on Noometry and DeepSeek-V2 (MoE-236B, May 2024) has 10.