Model comparison
DeepSeek LLM 67B vs Llama 3.2 1B
DeepSeek LLM 67B is the stronger model overall, scoring 24.9 to 20.1 on the Noometry Index.
Last verified . 14 shared benchmarks.
DeepSeek
MetaSummary
- They share 14 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 6 categories and Llama 3.2 1B in 2 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in coding, where DeepSeek LLM 67B leads 31.9 to 21.1.
Side by side
| DeepSeek LLM 67B | Llama 3.2 1B | |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 24.9 | 20.1 |
| Released | 2023-11-29 | 2024-09-24 |
| Weights | Open | Open |
| Context window | — | 60K |
| Max output | — | 54K |
| Input $ / M tokens | — | $0.027 |
| Output $ / M tokens | — | $0.20 |
| Results tracked | 15 | 22 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek LLM 67B leads
DeepSeek LLM 67B: 31.9 (#278), Llama 3.2 1B: 21.1 (#338)
| Benchmark | DeepSeek LLM 67B | Llama 3.2 1B |
|---|---|---|
| LMArena Coding | 1096 | 1070 |
| BigCodeBench Instruct | — | 8.2% |
| BigCodeBench Complete | — | 11.3% |
Agentic & Tool Use Not comparable
DeepSeek LLM 67B: —, Llama 3.2 1B: 14.6 (#150)
| Benchmark | DeepSeek LLM 67B | Llama 3.2 1B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 10.8% |
| BALROG | — | 6.6% |
Reasoning Too close to call
DeepSeek LLM 67B: 16.5 (#304), Llama 3.2 1B: 16.2 (#308)
| Benchmark | DeepSeek LLM 67B | Llama 3.2 1B |
|---|---|---|
| Chess Puzzles | 0% | 0% |
| LMArena Hard Prompts | 1070 | 1044 |
| Epoch Capabilities Index | 110.5 | 101.99 |
Math Llama 3.2 1B leads
DeepSeek LLM 67B: 8.7 (#324), Llama 3.2 1B: 10.4 (#313)
| Benchmark | DeepSeek LLM 67B | Llama 3.2 1B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | 0.6% |
| LMArena Math | 1108 | 1086 |
| MATH Level 5 | 6.4% | — |
Knowledge Too close to call
DeepSeek LLM 67B: 7.0 (#313), Llama 3.2 1B: 7.2 (#312)
| Benchmark | DeepSeek LLM 67B | Llama 3.2 1B |
|---|---|---|
| GPQA Diamond | 24.6% | 23.9% |
| LMArena Expert | — | 1007 |
Multilingual DeepSeek LLM 67B leads
DeepSeek LLM 67B: 29.4 (#267), Llama 3.2 1B: 23.8 (#292)
| Benchmark | DeepSeek LLM 67B | Llama 3.2 1B |
|---|---|---|
| LMArena Non-English | 1073 | 973 |
| LMArena Chinese | 1132 | 959 |
| LMArena German | — | 1014 |
| LMArena Russian | — | 941 |
Instruction Following DeepSeek LLM 67B leads
DeepSeek LLM 67B: 55.4 (#277), Llama 3.2 1B: 52.4 (#290)
| Benchmark | DeepSeek LLM 67B | Llama 3.2 1B |
|---|---|---|
| LMArena Instruction Following | 1079 | 1031 |
Long Context DeepSeek LLM 67B leads
DeepSeek LLM 67B: 33.1 (#265), Llama 3.2 1B: 31.9 (#274)
| Benchmark | DeepSeek LLM 67B | Llama 3.2 1B |
|---|---|---|
| LMArena Longer Query | 1092 | 1050 |
Writing & Preference DeepSeek LLM 67B leads
DeepSeek LLM 67B: 31.6 (#282), Llama 3.2 1B: 21.3 (#310)
| Benchmark | DeepSeek LLM 67B | Llama 3.2 1B |
|---|---|---|
| LMArena Text | 1105 | 1055 |
| LMArena Creative Writing | 1067 | 1033 |
| LMArena Multi-Turn | 1082 | 1030 |
| EQ-Bench Creative Writing | — | 200 |
Frequently asked questions
Is DeepSeek LLM 67B better than Llama 3.2 1B?
DeepSeek LLM 67B is the stronger model overall, scoring 24.9 to 20.1 on the Noometry Index.
Is DeepSeek LLM 67B or Llama 3.2 1B better for coding?
DeepSeek LLM 67B scores higher on coding benchmarks: 31.9 versus 21.1 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and Llama 3.2 1B share?
14 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Llama 3.2 1B has 22.