Model comparison
DeepSeek-V3.1 vs Llama 13b
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 24.4 on the Noometry Index.
Last verified . 9 shared benchmarks.
DeepSeek
MetaSummary
- They share 9 benchmarks with published results for both. DeepSeek-V3.1 scores higher in 6 categories and Llama 13b in 0 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where DeepSeek-V3.1 leads 60.3 to 13.8.
Side by side
| DeepSeek-V3.1 | Llama 13b | |
|---|---|---|
| Provider | DeepSeek | Meta |
| Noometry Index | 42.8 | 24.4 |
| Released | 2025-08-21 | 2023-02-24 |
| Weights | Open | Open |
| Context window | 164K | — |
| Max output | 8K | — |
| Input $ / M tokens | $0.25 | — |
| Output $ / M tokens | $0.95 | — |
| Results tracked | 27 | 21 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding DeepSeek-V3.1 leads
DeepSeek-V3.1: 40.3 (#144), Llama 13b: 21.4 (#337)
| Benchmark | DeepSeek-V3.1 | Llama 13b |
|---|---|---|
| LMArena Coding | 1417 | 683 |
| WeirdML | 38.4% | — |
Reasoning DeepSeek-V3.1 leads
DeepSeek-V3.1: 27.9 (#110), Llama 13b: 14.0 (#329)
| Benchmark | DeepSeek-V3.1 | Llama 13b |
|---|---|---|
| LMArena Hard Prompts | 1417 | 728 |
| Epoch Capabilities Index | 139.92 | 100.58 |
| SimpleBench | 40% | — |
| Kagi LLM Benchmark | 53.2% | — |
| DTBench | 82.7% | — |
| LMCA | 24.3% | — |
| BIG-Bench Hard | — | 37.9% |
| ForecastBench | 58 | — |
| HellaSwag | — | 79.2% |
| LAMBADA | — | 75.2% |
| PIQA | — | 80.1% |
| WinoGrande | — | 73% |
Math DeepSeek-V3.1 leads
DeepSeek-V3.1: 38.9 (#122), Llama 13b: 26.7 (#256)
| Benchmark | DeepSeek-V3.1 | Llama 13b |
|---|---|---|
| LMArena Math | 1420 | 838 |
| GSM8K | — | 20.6% |
Knowledge Not comparable
DeepSeek-V3.1: 43.7 (#90), Llama 13b: —
| Benchmark | DeepSeek-V3.1 | Llama 13b |
|---|---|---|
| Vectara Hallucination Rate | 5.5% | — |
| LMArena Expert | 1405 | — |
| ARC (AI2) Challenge | — | 52.7% |
| BoolQ | — | 78.7% |
| MMLU | — | 47.7% |
| OpenBookQA | — | 56.4% |
| TriviaQA | — | 77.9% |
Multimodal Not comparable
DeepSeek-V3.1: —, Llama 13b: —
| Benchmark | DeepSeek-V3.1 | Llama 13b |
|---|---|---|
| ScienceQA | — | 43.3% |
Multilingual DeepSeek-V3.1 leads
DeepSeek-V3.1: 51.6 (#106), Llama 13b: 16.6 (#297)
| Benchmark | DeepSeek-V3.1 | Llama 13b |
|---|---|---|
| LMArena Non-English | 1400 | 819 |
| LMArena Chinese | 1469 | — |
| LMArena French | 1447 | — |
| LMArena German | 1411 | — |
| LMArena Japanese | 1378 | — |
| LMArena Korean | 1337 | — |
| LMArena Russian | 1405 | — |
| LMArena Spanish | 1431 | — |
Instruction Following DeepSeek-V3.1 leads
DeepSeek-V3.1: 73.9 (#110), Llama 13b: 36.7 (#305)
| Benchmark | DeepSeek-V3.1 | Llama 13b |
|---|---|---|
| LMArena Instruction Following | 1400 | 781 |
Long Context Not comparable
DeepSeek-V3.1: 36.3 (#232), Llama 13b: —
| Benchmark | DeepSeek-V3.1 | Llama 13b |
|---|---|---|
| Fiction.LiveBench | 52.8% | — |
| LMArena Longer Query | 1422 | — |
Writing & Preference DeepSeek-V3.1 leads
DeepSeek-V3.1: 60.3 (#98), Llama 13b: 13.8 (#312)
| Benchmark | DeepSeek-V3.1 | Llama 13b |
|---|---|---|
| LMArena Text | 1420 | 834 |
| LMArena Creative Writing | 1401 | 794 |
| LMArena Multi-Turn | 1408 | 753 |
| EQ-Bench Creative Writing | 1436 | — |
Frequently asked questions
Is DeepSeek-V3.1 better than Llama 13b?
DeepSeek-V3.1 is the stronger model overall, scoring 42.8 to 24.4 on the Noometry Index.
Is DeepSeek-V3.1 or Llama 13b better for coding?
DeepSeek-V3.1 scores higher on coding benchmarks: 40.3 versus 21.4 in the Noometry coding category.
How many benchmarks do DeepSeek-V3.1 and Llama 13b share?
9 benchmarks have published results for both models. DeepSeek-V3.1 has 27 scored results on Noometry and Llama 13b has 21.