Model comparison
DeepSeek LLM 67B vs Llama 3.1-70B
Llama 3.1-70B is the stronger model overall, scoring 29.6 to 24.9 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 1 category and Llama 3.1-70B in 7 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Llama 3.1-70B leads 24.2 to 7.0.
- The biggest single-benchmark swing is MATH Level 5: 6.4% for DeepSeek LLM 67B and 36.7% for Llama 3.1-70B.
Side by side
| DeepSeek LLM 67B | Llama 3.1-70B | |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 24.9 | 29.6 |
| Released | 2023-11-29 | 2024-07-23 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.40 |
| Output $ / M tokens | — | $0.40 |
| Results tracked | 15 | 35 |
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 3.1-70B: 30.3 (#296)
| Benchmark | DeepSeek LLM 67B | Llama 3.1-70B |
|---|---|---|
| LMArena Coding | 1096 | 1260 |
| WeirdML | — | 9% |
| BigCodeBench Instruct | — | 46.1% |
| BigCodeBench Complete | — | 54.8% |
Agentic & Tool Use Not comparable
DeepSeek LLM 67B: —, Llama 3.1-70B: 25.1 (#112)
| Benchmark | DeepSeek LLM 67B | Llama 3.1-70B |
|---|---|---|
| TheAgentCompany | — | 6.9% |
| BALROG | — | 27.9% |
Reasoning Llama 3.1-70B leads
DeepSeek LLM 67B: 16.5 (#304), Llama 3.1-70B: 21.6 (#220)
| Benchmark | DeepSeek LLM 67B | Llama 3.1-70B |
|---|---|---|
| LMArena Hard Prompts | 1070 | 1241 |
| Epoch Capabilities Index | 110.5 | 125.92 |
| Chess Puzzles | 0% | — |
| DTBench | — | 60% |
| LMCA | — | 14.8% |
Math Llama 3.1-70B leads
DeepSeek LLM 67B: 8.7 (#324), Llama 3.1-70B: 13.5 (#304)
| Benchmark | DeepSeek LLM 67B | Llama 3.1-70B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | 3.6% |
| LMArena Math | 1108 | 1252 |
| MATH Level 5 | 6.4% | 36.7% |
| Omni-MATH | — | 21% |
Knowledge Llama 3.1-70B leads
DeepSeek LLM 67B: 7.0 (#313), Llama 3.1-70B: 24.2 (#269)
| Benchmark | DeepSeek LLM 67B | Llama 3.1-70B |
|---|---|---|
| GPQA Diamond | 24.6% | 44.2% |
| MMLU-Pro | — | 65.3% |
| GPQA (HELM) | — | 42.6% |
| LMArena Expert | — | 1209 |
| MMLU | — | 80.1% |
Multilingual Llama 3.1-70B leads
DeepSeek LLM 67B: 29.4 (#267), Llama 3.1-70B: 38.8 (#225)
| Benchmark | DeepSeek LLM 67B | Llama 3.1-70B |
|---|---|---|
| LMArena Non-English | 1073 | 1219 |
| LMArena Chinese | 1132 | 1215 |
| LMArena French | — | 1261 |
| LMArena German | — | 1222 |
| LMArena Japanese | — | 1132 |
| LMArena Korean | — | 1140 |
| LMArena Russian | — | 1234 |
| LMArena Spanish | — | 1253 |
Instruction Following Llama 3.1-70B leads
DeepSeek LLM 67B: 55.4 (#277), Llama 3.1-70B: 65.3 (#223)
| Benchmark | DeepSeek LLM 67B | Llama 3.1-70B |
|---|---|---|
| LMArena Instruction Following | 1079 | 1231 |
| IFEval | — | 82.1% |
Long Context Llama 3.1-70B leads
DeepSeek LLM 67B: 33.1 (#265), Llama 3.1-70B: 37.6 (#214)
| Benchmark | DeepSeek LLM 67B | Llama 3.1-70B |
|---|---|---|
| LMArena Longer Query | 1092 | 1241 |
Writing & Preference Llama 3.1-70B leads
DeepSeek LLM 67B: 31.6 (#282), Llama 3.1-70B: 35.4 (#267)
| Benchmark | DeepSeek LLM 67B | Llama 3.1-70B |
|---|---|---|
| LMArena Text | 1105 | 1261 |
| LMArena Creative Writing | 1067 | 1232 |
| LMArena Multi-Turn | 1082 | 1256 |
| EQ-Bench Creative Writing | — | 784 |
| WildBench | — | 75.8% |
Frequently asked questions
Is DeepSeek LLM 67B better than Llama 3.1-70B?
Llama 3.1-70B is the stronger model overall, scoring 29.6 to 24.9 on the Noometry Index.
Is DeepSeek LLM 67B or Llama 3.1-70B better for coding?
DeepSeek LLM 67B scores higher on coding benchmarks: 31.9 versus 30.3 in the Noometry coding category.
How many benchmarks do DeepSeek LLM 67B and Llama 3.1-70B share?
14 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Llama 3.1-70B has 35.