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 | 2024-03-11 |
| Parameters (B) | 1.3 |
| Architecture | Vision-Language Transformer (SigLIP-L + DeepSeek-LLM) |
| Context Window | 4096 |
| Input Modalities | text, image |
| Open Source | Yes |
| License | other |
| Score | 0 |
| VRAM (GB) | 4 |
| Compute | Single GPU with ~4GB VRAM for bf16/fp16 inference; suitable for consumer GPUs and edge/mobile deployment |
| Benchmark: AGIEval | 14 |
| Benchmark: GSM8K | 18 |
| Benchmark: HellaSwag | 56 |
| Benchmark: MBPP | 10 |
| Benchmark: MMLU | 32.5 |
| Benchmark: MMMU | 32.2 |
| Pricing (Input) | — |
| Pricing (Output) | — |