Can Qwen3 Coder 30B A3B Instruct run on the RX 7700 XT?
No. Even at 2-bit with 8K context, Qwen3 Coder 30B A3B Instruct needs about 11.9 GB, more than the 11.4 GB usable on the RX 7700 XT.
Memory by quant and context
| Quant | 4K ctx | 8K ctx | 32K ctx | 128K ctx | Max tok/s (est., 8K) |
|---|---|---|---|---|---|
| 2-bit (Q2_K) • | 11.6 GB No | 11.9 GB No | 14.2 GB No | 23.2 GB No | — |
| 3-bit (Q3_K_M) • | 14.8 GB No | 15.1 GB No | 17.4 GB No | 26.4 GB No | — |
| 4-bit (Q4_K_M) • | 18.3 GB No | 18.7 GB No | 21.0 GB No | 30.0 GB No | — |
| 5-bit (Q5_K_M) • | 21.3 GB No | 21.7 GB No | 23.9 GB No | 32.9 GB No | — |
| 6-bit (Q6_K) • | 24.4 GB No | 24.8 GB No | 27.1 GB No | 36.1 GB No | — |
| 8-bit (Q8_0) • | 31.3 GB No | 31.7 GB No | 33.9 GB No | 42.9 GB No | — |
| 16-bit (F16/BF16) • | 58.0 GB No | 58.3 GB No | 60.6 GB No | 69.6 GB No | — |
• real GGUF file size from unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF. Speed is a bandwidth ceiling (432 GB/s), not a benchmark. Methodology.
Hardware that does fit it
RTX 4060 Ti 16GBRTX 4070 Ti SuperRTX 4080RTX 4080 SuperRTX 5060 Ti 16GBRTX 5070 Ti
More
Qwen3 Coder 30B A3B Instruct on every GPU and Mac · Every model that runs on the RX 7700 XT · Other models that fit 12 GB VRAM · Try other settings in the calculator
Data: Hugging Face Hub (Qwen/Qwen3-Coder-30B-A3B-Instruct), AMD specs. Data as of 2026-11-17.