Model comparison
DeepSeek-V2 (MoE-236B, May 2024) vs Llama 3.2 1B
Llama 3.2 1B has enough public results to be ranked (#354); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Last verified . 3 shared benchmarks.
DeepSeek
MetaSummary
- They share 3 benchmarks with published results for both. DeepSeek-V2 (MoE-236B, May 2024) scores higher in 1 category and Llama 3.2 1B 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 21.1.
- The biggest single-benchmark swing is BigCodeBench Complete: 59.4% for DeepSeek-V2 (MoE-236B, May 2024) and 11.3% for Llama 3.2 1B.
Side by side
| DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.2 1B | |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 40.3 | 20.1 |
| Released | 2024-05-07 | 2024-09-24 |
| Weights | Open | Open |
| Context window | — | 60K |
| Max output | — | 54K |
| Input $ / M tokens | — | $0.027 |
| Output $ / M tokens | — | $0.20 |
| Results tracked | 10 | 22 |
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 3.2 1B: 21.1 (#338)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.2 1B |
|---|---|---|
| BigCodeBench Instruct | 48.9% | 8.2% |
| BigCodeBench Complete | 59.4% | 11.3% |
| LMArena Coding | — | 1070 |
Agentic & Tool Use Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Llama 3.2 1B: 14.6 (#150)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.2 1B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 10.8% |
| BALROG | — | 6.6% |
Reasoning Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Llama 3.2 1B: 16.2 (#308)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.2 1B |
|---|---|---|
| Epoch Capabilities Index | 124.77 | 101.99 |
| Chess Puzzles | — | 0% |
| LMArena Hard Prompts | — | 1044 |
| BIG-Bench Hard | 78.8% | — |
| HellaSwag | 87.1% | — |
| PIQA | 83.9% | — |
| WinoGrande | 86.3% | — |
Math Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Llama 3.2 1B: 10.4 (#313)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.2 1B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 0.6% |
| LMArena Math | — | 1086 |
Knowledge Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Llama 3.2 1B: 7.2 (#312)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.2 1B |
|---|---|---|
| GPQA Diamond | — | 23.9% |
| LMArena Expert | — | 1007 |
| ARC (AI2) Challenge | 92.2% | — |
| MMLU | 78.4% | — |
| TriviaQA | 80% | — |
Multilingual Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Llama 3.2 1B: 23.8 (#292)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.2 1B |
|---|---|---|
| LMArena Non-English | — | 973 |
| LMArena Chinese | — | 959 |
| LMArena German | — | 1014 |
| LMArena Russian | — | 941 |
Instruction Following Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Llama 3.2 1B: 52.4 (#290)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.2 1B |
|---|---|---|
| LMArena Instruction Following | — | 1031 |
Long Context Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Llama 3.2 1B: 31.9 (#274)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.2 1B |
|---|---|---|
| LMArena Longer Query | — | 1050 |
Writing & Preference Not comparable
DeepSeek-V2 (MoE-236B, May 2024): —, Llama 3.2 1B: 21.3 (#310)
| Benchmark | DeepSeek-V2 (MoE-236B, May 2024) | Llama 3.2 1B |
|---|---|---|
| LMArena Text | — | 1055 |
| LMArena Creative Writing | — | 1033 |
| EQ-Bench Creative Writing | — | 200 |
| LMArena Multi-Turn | — | 1030 |
Frequently asked questions
Is DeepSeek-V2 (MoE-236B, May 2024) better than Llama 3.2 1B?
Llama 3.2 1B has enough public results to be ranked (#354); DeepSeek-V2 (MoE-236B, May 2024) does not yet, so treat this comparison as directional.
Is DeepSeek-V2 (MoE-236B, May 2024) or Llama 3.2 1B better for coding?
DeepSeek-V2 (MoE-236B, May 2024) scores higher on coding benchmarks: 40.4 versus 21.1 in the Noometry coding category.
How many benchmarks do DeepSeek-V2 (MoE-236B, May 2024) and Llama 3.2 1B share?
3 benchmarks have published results for both models. DeepSeek-V2 (MoE-236B, May 2024) has 10 scored results on Noometry and Llama 3.2 1B has 22.