// DeepSeek · Vision-language
deepseek-vl2-tiny: VRAM & local deployment
3.37B total · 1B active (MoE) parameters. About 3.2 GB at 4-bit and 8.4 GB at BF16 with one 8K-token request.
released 2024-12-13licence: Commercial with conditionsmemory figures are estimates
01Specs
| Publisher | DeepSeek · deepseek-ai |
|---|---|
| Family / series | DeepSeek · deepseek-vl2-tiny |
| Type | Vision-language (image-text-to-text) |
| Parameters | 3.37B total · 1B active (MoE) Source: counted from official safetensors weightsActive parameters from the model card: “DeepSeek-VL2-Tiny, DeepSeek-VL2-Small and DeepSeek-VL2, with 1.0B, 2.8B and 4.5B activated parameters respectively.” |
| Context length | 4K tokens (4,096)max_position_embeddings in config.json |
| Architecture | 12 layers · GQA · MoE |
| Released weights | BF16 · 6.7 GBdownload |
| Licence | deepseek · Commercial with conditionsCommercial use allowed under the publisher’s own licence terms (user caps, acceptable-use policy, attribution or region limits). Read it before shipping. |
| Released | 2024-12-13 (Hugging Face repo created) |
02How much VRAM it needs (estimate)
Total = weights + 10% + 0.5GB overhead + KV cache for one request (FP16 KV). More concurrent users multiply the KV column.
| Precision | Weights | Total @ 8K | Smallest GPU | Mac |
|---|---|---|---|---|
| BF16 / FP16 (original) | 6.7 GB | 8.4 GB | 12GB GPU | Mac 16GB |
| FP8 / INT8 / Q8_0 | 3.6 GB | 4.9 GB | 8GB GPU | Mac 16GB |
| 4-bit (Q4_K_M / AWQ / GPTQ-Int4) | 2.0 GB | 3.2 GB | 8GB GPU | Mac 16GB |
KV cache per request: 8K 0.5 GB
Bits per weight: BF16 16, FP8 8.5, 4-bit 4.85. “Fits” = total ≤ 90% of memory; Mac GPU can use ~2/3 of unified memory (≤36GB) or ~3/4 (larger). For MoE, all experts must be in memory, so size by total parameters. Method →
03How to run it
Only runtimes the publisher or the runtime’s official library names for this model.
- Transformers — mentioned in the official model card
04Official repositories & quantised builds
1 official repo(s) from deepseek-ai.
- deepseek-vl2-tiny main
05Other DeepSeek models
All 34 →| Model | Params | 4-bit total | Released |
|---|---|---|---|
| DeepSeek-V4.1-Flash-1017 | 748B / 16B | 499 GB | 2026-10-17 |
| DeepSeek-V4.1-Flash | 748B / 16B | 499 GB | 2026-09-10 |
| DeepSeek-V4-Pro-0813 | 1.65T | 1.10 TB | 2026-08-13 |
| DeepSeek-V4-Flash-0731 | 304.2B | 203 GB | 2026-07-31 |
| DeepSeek-V4-Flash | 284B / 13B | 190 GB | 2026-04-22 |
| DeepSeek-V4-Pro | 1.6T / 49B | 1.07 TB | 2026-04-22 |
| DeepSeek-OCR-2 | 3.39B | 3.3 GB | 2026-01-27 |
| DeepSeek-V3.2 | 685.4B | 458 GB | 2025-12-01 |
06Sources
Data checked 2024-12-18. Parameters, context and licence are copied from the official repository; memory figures are our estimates.