# DeepSeek LLM 67B vs Llama 3.1-8B

> DeepSeek LLM 67B is the stronger model overall, scoring 24.9 to 23.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/deepseek-llm-67b-vs-llama-3-1-8b
- Last updated: 2026-10-10
- Shared benchmarks: 15

## Summary

- They share 15 benchmarks with published results for both. DeepSeek LLM 67B scores higher in 3 categories and Llama 3.1-8B in 5 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in coding, where DeepSeek LLM 67B leads 31.9 to 20.2.
- The biggest single-benchmark swing is MATH Level 5: 6.4% for DeepSeek LLM 67B and 22.9% for Llama 3.1-8B.

## Snapshot

| | DeepSeek LLM 67B | Llama 3.1-8B |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 24.9 | 23.0 |
| Rank | 347 | 352 |
| Context | — | 128K |
| Input $/M | — | $0.05 |
| Output $/M | — | $0.08 |
| Weights | Open | Open |

## Coding

- DeepSeek LLM 67B: 31.9 (#278)
- Llama 3.1-8B: 20.2 (#340)

| Benchmark | DeepSeek LLM 67B | Llama 3.1-8B |
|---|---|---|
| LMArena Coding | 1096 | 1195 |
| SciCode | — | 13.2% |
| WeirdML | — | 1.7% |
| BigCodeBench Instruct | — | 32.8% |
| BigCodeBench Complete | — | 40.5% |
| HumanEval+ | — | 62.8% |
| MBPP+ | — | 55.6% |

## Agentic & Tool Use

- DeepSeek LLM 67B: —
- Llama 3.1-8B: 22.5 (#131)

| Benchmark | DeepSeek LLM 67B | Llama 3.1-8B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 25.8% |
| BALROG | — | 15.1% |

## Reasoning

- DeepSeek LLM 67B: 16.5 (#304)
- Llama 3.1-8B: 14.9 (#321)

| Benchmark | DeepSeek LLM 67B | Llama 3.1-8B |
|---|---|---|
| Chess Puzzles | 0% | 0% |
| LMArena Hard Prompts | 1070 | 1175 |
| Epoch Capabilities Index | 110.5 | 116.57 |
| CritPt | — | 0% |
| DTBench | — | 50.9% |
| LMCA | — | 5.4% |
| PIQA | — | 81.2% |

## Math

- DeepSeek LLM 67B: 8.7 (#324)
- Llama 3.1-8B: 10.2 (#317)

| Benchmark | DeepSeek LLM 67B | Llama 3.1-8B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 0.8% | 1.7% |
| LMArena Math | 1108 | 1179 |
| MATH Level 5 | 6.4% | 22.9% |
| Omni-MATH | — | 13.7% |
| GSM8K | — | 82.4% |

## Knowledge

- DeepSeek LLM 67B: 7.0 (#313)
- Llama 3.1-8B: 8.0 (#307)

| Benchmark | DeepSeek LLM 67B | Llama 3.1-8B |
|---|---|---|
| GPQA Diamond | 24.6% | 27% |
| MMLU-Pro | — | 40.6% |
| GPQA (HELM) | — | 24.7% |
| LMArena Expert | — | 1144 |
| BoolQ | — | 82.8% |
| MMLU | — | 56.1% |

## Multilingual

- DeepSeek LLM 67B: 29.4 (#267)
- Llama 3.1-8B: 34.0 (#249)

| Benchmark | DeepSeek LLM 67B | Llama 3.1-8B |
|---|---|---|
| LMArena Non-English | 1073 | 1148 |
| LMArena Chinese | 1132 | 1151 |
| LMArena French | — | 1177 |
| LMArena German | — | 1144 |
| LMArena Japanese | — | 1061 |
| LMArena Korean | — | 1053 |
| LMArena Russian | — | 1158 |
| LMArena Spanish | — | 1169 |

## Instruction Following

- DeepSeek LLM 67B: 55.4 (#277)
- Llama 3.1-8B: 58.9 (#258)

| Benchmark | DeepSeek LLM 67B | Llama 3.1-8B |
|---|---|---|
| LMArena Instruction Following | 1079 | 1159 |
| IFEval | — | 74.3% |

## Long Context

- DeepSeek LLM 67B: 33.1 (#265)
- Llama 3.1-8B: 35.8 (#238)

| Benchmark | DeepSeek LLM 67B | Llama 3.1-8B |
|---|---|---|
| LMArena Longer Query | 1092 | 1182 |

## Writing & Preference

- DeepSeek LLM 67B: 31.6 (#282)
- Llama 3.1-8B: 29.7 (#290)

| Benchmark | DeepSeek LLM 67B | Llama 3.1-8B |
|---|---|---|
| LMArena Text | 1105 | 1187 |
| LMArena Creative Writing | 1067 | 1154 |
| LMArena Multi-Turn | 1082 | 1172 |
| EQ-Bench Creative Writing | — | 713 |
| WildBench | — | 68.7% |

## FAQ

### Is DeepSeek LLM 67B better than Llama 3.1-8B?

DeepSeek LLM 67B is the stronger model overall, scoring 24.9 to 23.0 on the Noometry Index.

### Is DeepSeek LLM 67B or Llama 3.1-8B better for coding?

DeepSeek LLM 67B scores higher on coding benchmarks: 31.9 versus 20.2 in the Noometry coding category.

### How many benchmarks do DeepSeek LLM 67B and Llama 3.1-8B share?

15 benchmarks have published results for both models. DeepSeek LLM 67B has 15 scored results on Noometry and Llama 3.1-8B has 43.