Model comparison
DeepSeek-V2 (MoE-236B, May 2024) vs Llama 3.1-70B
Llama 3.1-70B has enough public results to be ranked (#308); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Last verified . 4 shared benchmarks.
DeepSeek
MetaSummary
- They share 4 benchmarks with published results for both. DeepSeek-V2 (MoE-236B, May 2024) scores higher in 1 category and Llama 3.1-70B in 0 categories; one gap is clear of the uncertainty.
- The widest gap is in coding, where DeepSeek-V2 (MoE-236B, May 2024) leads 40.4 to 30.3.
Side by side
| DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.1-70B | |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 40.3 | 29.6 |
| Released | 2024-05-07 | 2024-07-23 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.40 |
| Output $ / M tokens | — | $0.40 |
| Results tracked | 10 | 35 |
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): 40.4 (#139), Llama 3.1-70B: 30.3 (#296)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.1-70B |
|---|---|---|
| BigCodeBench Instruct | 48.9% | 46.1% |
| BigCodeBench Complete | 59.4% | 54.8% |
| WeirdML | — | 9% |
| LMArena Coding | — | 1260 |
Agentic & Tool Use Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Llama 3.1-70B: 25.1 (#112)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.1-70B |
|---|---|---|
| TheAgentCompany | — | 6.9% |
| BALROG | — | 27.9% |
Reasoning Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Llama 3.1-70B: 21.6 (#220)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.1-70B |
|---|---|---|
| Epoch Capabilities Index | 124.77 | 125.92 |
| LMArena Hard Prompts | — | 1241 |
| DTBench | — | 60% |
| LMCA | — | 14.8% |
| BIG-Bench Hard | 78.8% | — |
| HellaSwag | 87.1% | — |
| PIQA | 83.9% | — |
| WinoGrande | 86.3% | — |
Math Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Llama 3.1-70B: 13.5 (#304)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.1-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 3.6% |
| Omni-MATH | — | 21% |
| LMArena Math | — | 1252 |
| MATH Level 5 | — | 36.7% |
Knowledge Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Llama 3.1-70B: 24.2 (#269)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.1-70B |
|---|---|---|
| MMLU | 78.4% | 80.1% |
| GPQA Diamond | — | 44.2% |
| MMLU-Pro | — | 65.3% |
| GPQA (HELM) | — | 42.6% |
| LMArena Expert | — | 1209 |
| ARC (AI2) Challenge | 92.2% | — |
| TriviaQA | 80% | — |
Multilingual Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Llama 3.1-70B: 38.8 (#225)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.1-70B |
|---|---|---|
| LMArena Non-English | — | 1219 |
| LMArena Chinese | — | 1215 |
| LMArena French | — | 1261 |
| LMArena German | — | 1222 |
| LMArena Japanese | — | 1132 |
| LMArena Korean | — | 1140 |
| LMArena Russian | — | 1234 |
| LMArena Spanish | — | 1253 |
Instruction Following Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Llama 3.1-70B: 65.3 (#223)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.1-70B |
|---|---|---|
| IFEval | — | 82.1% |
| LMArena Instruction Following | — | 1231 |
Long Context Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Llama 3.1-70B: 37.6 (#214)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.1-70B |
|---|---|---|
| LMArena Longer Query | — | 1241 |
Writing & Preference Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Llama 3.1-70B: 35.4 (#267)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.1-70B |
|---|---|---|
| LMArena Text | — | 1261 |
| LMArena Creative Writing | — | 1232 |
| EQ-Bench Creative Writing | — | 784 |
| WildBench | — | 75.8% |
| LMArena Multi-Turn | — | 1256 |
Frequently asked questions
Is DeepSeek-V2 (MoE-236B, May 2024) better than Llama 3.1-70B?
Llama 3.1-70B has enough public results to be ranked (#308); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Is DeepSeek-V2 (MoE-236B, May 2024) or Llama 3.1-70B better for coding?
DeepSeek-V2 (MoE-236B, May 2024) scores higher on coding benchmarks: 40.4 versus 30.3 in the Noometry coding category.
How many benchmarks do DeepSeek-V2 (MoE-236B, May 2024) and Llama 3.1-70B share?
4 benchmarks have published results for both models. DeepSeek-V2 (MoE-236B, May 2024) has 10 scored results on Noometry and Llama 3.1-70B has 35.