Compare LLM Models
Select 2-5 models to compare side by side.
Select models to compare
0 / 5 selected
Top downloads
Select 2-5 models
| Attribute |
deepseek-ai
|
|---|---|
| Company | deepseek-ai |
| Release Date | 2023-10-28 |
| Parameters (B) | 1.3 |
| Architecture | Decoder-only Transformer |
| Context Window | 16384 |
| Input Modalities | text |
| Open Source | Yes |
| License | other |
| Score | 0 |
| VRAM (GB) | 3.3 |
| Compute | ~3.3GB VRAM at BF16 (2.6GB weights); Q4_K_M quantization runs on ~1.5GB VRAM; full 16K context adds ~2.8GB KV cache |
| Benchmark: GSM8K | 14.6 |
| Benchmark: HumanEval | 34.8 |
| Benchmark: MATH | 16.8 |
| Benchmark: MBPP | 46.2 |
| Pricing (Input) | — |
| Pricing (Output) | — |