Can I run Qwen3-Coder-30B-A3B-Instruct-GGUF on an RTX 5080?
Yes, just
Its UD-Q3_K_XL file (13.8 GB) fits in the 14.4 GB an RTX 5080 can use, with room for ≈ 6K tokens of context.
Full analysis of Qwen3-Coder-30B-A3B-Instruct-GGUF
Everything that runs on RTX 5080
Get the UD-Q3_K_XL file on Hugging Face
ollama run hf.co/unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF:UD-Q3_K_XLFiles and context on RTX 5080
Each file of Qwen3-Coder-30B-A3B-Instruct-GGUF (and of GGUF versions the site has analyzed), and how much context fits next to it, with the context cache in 16-bit or 8-bit.
| File | Size | On RTX 5080 | Context (16-bit) | Context (8-bit) |
|---|---|---|---|---|
| BF16 | 61.1 GB | Doesn't fit | — | — |
| UD-Q8_K_XL | 36.0 GB | Doesn't fit | — | — |
| Q8_0 | 32.5 GB | Doesn't fit | — | — |
| UD-Q6_K_XL | 26.3 GB | Doesn't fit | — | — |
| Q6_K | 25.1 GB | Doesn't fit | — | — |
| UD-Q5_K_XL | 21.7 GB | Doesn't fit | — | — |
| Q5_K_M | 21.7 GB | Doesn't fit | — | — |
| Q5_K_S | 21.1 GB | Doesn't fit | — | — |
| Q4_1 | 19.2 GB | Doesn't fit | — | — |
| Q4_K_M | 18.6 GB | Doesn't fit | — | — |
| UD-Q4_K_XL | 17.7 GB | Doesn't fit | — | — |
| Q4_K_S | 17.5 GB | Doesn't fit | — | — |
| Q4_0 | 17.4 GB | Doesn't fit | — | — |
| IQ4_NL | 17.3 GB | Doesn't fit | — | — |
| IQ4_XS | 16.4 GB | Doesn't fit | — | — |
| Q3_K_M | 14.7 GB | Doesn't fit | — | — |
| UD-Q3_K_XL | 13.8 GB | Tight | 6K | 12K |
| Q3_K_S | 13.3 GB | Tight | 11K | 22K |
| UD-IQ3_XXS | 12.8 GB | Tight | 15K | 31K |
| UD-Q2_K_XL | 11.8 GB | Tight | 26K | 52K |
| Q2_K_L | 11.3 GB | Fits | 30K | 61K |
| Q2_K | 11.3 GB | Fits | 31K | 62K |
| UD-IQ2_M | 10.8 GB | Fits | 35K | 71K |
| UD-IQ2_XXS | 10.3 GB | Fits | 40K | 81K |
| UD-IQ1_M | 9.63 GB | Fits | 47K | 95K |
| UD-IQ1_S | 8.91 GB | Fits | 54K | 109K |
| UD-TQ1_0 | 8.01 GB | Fits | 64K | 127K |