Can Moonlight 16B A3B Instruct run on the RX 7900 XT?
Yes. At 8K context the best fit is Q8_0, using about 16.8 GB of the 19.0 GB usable on the RX 7900 XT. Estimated speed ceiling: ~251 tokens/sec.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | Max tok/s (est., 8K) |
|---|---|---|---|
| 2-bit (Q2_K) • | 7.0 GB Fits | 7.1 GB Fits | 647 |
| 3-bit (Q3_K_M) • | 8.5 GB Fits | 8.7 GB Fits | 514 |
| 4-bit (Q4_K_M) • | 10.6 GB Fits | 10.8 GB Fits | 404 |
| 5-bit (Q5_K_M) • | 12.0 GB Fits | 12.2 GB Fits | 353 |
| 6-bit (Q6_K) • | 14.1 GB Fits | 14.2 GB Fits | 298 |
| 8-bit (Q8_0) • | 16.6 GB Fits | 16.8 GB Fits | 251 |
| 16-bit (F16/BF16) | 30.6 GB No | 30.7 GB No | — |
• real GGUF file size from mmnga/Moonlight-16B-A3B-Instruct-gguf. Speed is a bandwidth ceiling (800 GB/s), not a benchmark. Methodology.
More
Moonlight 16B A3B Instruct 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 (moonshotai/Moonlight-16B-A3B-Instruct), AMD specs. Data as of 2026-11-17.