Can deepseek moe 16b base run on the RX 7900 XT?
Yes. At 8K context the best fit is Q6_K, using about 16.0 GB of the 19.0 GB usable on the RX 7900 XT. Estimated speed ceiling: ~316 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | Max tok/s (est., 8K) |
|---|---|---|
| 2-bit (Q2_K) • | 7.7 GB Fits | 691 |
| 3-bit (Q3_K_M) • | 9.4 GB Fits | 546 |
| 4-bit (Q4_K_M) • | 11.6 GB Fits | 427 |
| 5-bit (Q5_K_M) • | 13.0 GB Fits | 374 |
| 6-bit (Q6_K) • | 15.1 GB Fits | 316 |
| 8-bit (Q8_0) • | 17.7 GB Tight | 266 |
| 16-bit (F16/BF16) | 32.0 GB No | — |
• real GGUF file size from mradermacher/deepseek-moe-16b-base-GGUF. Speed is a bandwidth ceiling (800 GB/s), not a benchmark. Methodology.
More
deepseek moe 16b base on every GPU and Mac · Every model that runs on the RX 7900 XT · Other models that fit 24 GB VRAM · Try other settings in the calculator
Data: Hugging Face Hub (deepseek-ai/deepseek-moe-16b-base), AMD specs. Data as of 2026-11-17.