Can deepseek vl2 tiny run on the RX 7900 GRE?
Yes. At 8K context the best fit is F16, using about 7.4 GB of the 15.2 GB usable on the RX 7900 GRE. Estimated speed ceiling: ~245 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | Max tok/s (est., 8K) |
|---|---|---|
| 2-bit (Q2_K) | 2.1 GB Fits | 1307 |
| 3-bit (Q3_K_M) | 2.4 GB Fits | 1003 |
| 4-bit (Q4_K_M) | 2.8 GB Fits | 802 |
| 5-bit (Q5_K_M) | 3.1 GB Fits | 688 |
| 6-bit (Q6_K) | 3.5 GB Fits | 598 |
| 8-bit (Q8_0) | 4.2 GB Fits | 461 |
| 16-bit (F16/BF16) | 7.2 GB Fits | 245 |
• real GGUF file size. Speed is a bandwidth ceiling (576 GB/s), not a benchmark. Methodology.
More
deepseek vl2 tiny on every GPU and Mac · Every model that runs on the RX 7900 GRE · Other models that fit 16 GB VRAM · Try other settings in the calculator
Data: Hugging Face Hub (deepseek-ai/deepseek-vl2-tiny), AMD specs. Data as of 2026-11-17.