Model comparison
DeepSeek LLM 67B vs Llama-3.3-70B-Instruct
Llama-3.3-70B-Instruct is the stronger model overall, scoring 30.6 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 3 categories and Llama-3.3-70B-Instruct in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Llama-3.3-70B-Instruct leads 30.6 to 7.0.
- The biggest single-benchmark swing is MATH Level 5: 6.4% for DeepSeek LLM 67B and 41.6% for Llama-3.3-70B-Instruct.
Side by side
| DeepSeek LLM 67B | Llama-3.3-70B-Instruct | |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 24.9 | 30.6 |
| Released | 2023-11-29 | 2024-12-06 |
| Weights | Open | Open |
| Context window | — | 128K |
| Max output | — | 4K |
| Input $ / M tokens | — | $0.10 |
| Output $ / M tokens | — | $0.32 |
| Results tracked | 15 | 43 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Too close to call
DeepSeek LLM 67B: 31.9 (#278), Llama-3.3-70B-Instruct: 31.0 (#290)
| Benchmark | DeepSeek LLM 67B | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Coding | 1096 | 1268 |
| SciCode | — | 26% |
| WeirdML | — | 14.4% |
| BigCodeBench Instruct | — | 46.9% |
| LiveBench Coding | — | 36.6% |
| BigCodeBench Complete | — | 57.5% |
Agentic & Tool Use Not comparable
DeepSeek LLM 67B: —, Llama-3.3-70B-Instruct: 25.8 (#105)
| Benchmark | DeepSeek LLM 67B | Llama-3.3-70B-Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 31.9% |
| BALROG | — | 23% |
Reasoning DeepSeek LLM 67B leads
DeepSeek LLM 67B: 16.5 (#304), Llama-3.3-70B-Instruct: 14.1 (#327)
| Benchmark | DeepSeek LLM 67B | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Hard Prompts | 1070 | 1257 |
| Epoch Capabilities Index | 110.5 | 127.33 |
| SimpleBench | — | 19.9% |
| CritPt | — | 0% |
| Chess Puzzles | 0% | — |
| LiveBench Reasoning | — | 50.8% |
| DTBench | — | 59.5% |
| LiveBench Data Analysis | — | 49.5% |
| LMCA | — | 17.5% |
| ForecastBench | — | 58.6 |
| LiveBench | — | 50.2% |
Math Llama-3.3-70B-Instruct leads
DeepSeek LLM 67B: 8.7 (#324), Llama-3.3-70B-Instruct: 15.3 (#298)
| Benchmark | DeepSeek LLM 67B | Llama-3.3-70B-Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | 5.1% |
| LMArena Math | 1108 | 1267 |
| MATH Level 5 | 6.4% | 41.6% |
| LiveBench Math | — | 42.2% |
Knowledge Llama-3.3-70B-Instruct leads
DeepSeek LLM 67B: 7.0 (#313), Llama-3.3-70B-Instruct: 30.6 (#226)
| Benchmark | DeepSeek LLM 67B | Llama-3.3-70B-Instruct |
|---|---|---|
| GPQA Diamond | 24.6% | 47.4% |
| Confabulations | — | 22.8% |
| Vectara Hallucination Rate | — | 4.1% |
| LMArena Expert | — | 1225 |
| MMLU | — | 86.3% |
Multilingual Llama-3.3-70B-Instruct leads
DeepSeek LLM 67B: 29.4 (#267), Llama-3.3-70B-Instruct: 39.9 (#220)
| Benchmark | DeepSeek LLM 67B | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Non-English | 1073 | 1236 |
| LMArena Chinese | 1132 | 1217 |
| LMArena French | — | 1281 |
| LMArena German | — | 1251 |
| LMArena Japanese | — | 1150 |
| LMArena Korean | — | 1143 |
| LMArena Russian | — | 1252 |
| LMArena Spanish | — | 1270 |
Instruction Following Llama-3.3-70B-Instruct leads
DeepSeek LLM 67B: 55.4 (#277), Llama-3.3-70B-Instruct: 71.1 (#157)
| Benchmark | DeepSeek LLM 67B | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Instruction Following | 1079 | 1242 |
| LiveBench Instruction Following | — | 82.7% |
Long Context DeepSeek LLM 67B leads
DeepSeek LLM 67B: 33.1 (#265), Llama-3.3-70B-Instruct: 26.4 (#295)
| Benchmark | DeepSeek LLM 67B | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Longer Query | 1092 | 1256 |
| Fiction.LiveBench | — | 33.3% |
Writing & Preference Llama-3.3-70B-Instruct leads
DeepSeek LLM 67B: 31.6 (#282), Llama-3.3-70B-Instruct: 47.6 (#207)
| Benchmark | DeepSeek LLM 67B | Llama-3.3-70B-Instruct |
|---|---|---|
| LMArena Text | 1105 | 1274 |
| LMArena Creative Writing | 1067 | 1250 |
| LMArena Multi-Turn | 1082 | 1280 |
| LiveBench Language | — | 39.2% |
Frequently asked questions
Is DeepSeek LLM 67B better than Llama-3.3-70B-Instruct?
Llama-3.3-70B-Instruct is the stronger model overall, scoring 30.6 to 24.9 on the Noometry Index.
Is DeepSeek LLM 67B or Llama-3.3-70B-Instruct better for coding?
They score almost the same on coding (31.9 vs 31.0); test both on your own repository before choosing.
How many benchmarks do DeepSeek LLM 67B and Llama-3.3-70B-Instruct share?
14 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Llama-3.3-70B-Instruct has 43.