Can I run Qwen3.6-35B-A3B-MTP-GGUF on an RTX 4090?
Yes, just
Its UD-Q4_K_S file (21.4 GB) fits in the 21.6 GB an RTX 4090 can use, with room for ≈ 7K tokens of context.
Full analysis of Qwen3.6-35B-A3B-MTP-GGUF
Everything that runs on RTX 4090
Get the UD-Q4_K_S file on Hugging Face
ollama run hf.co/unsloth/Qwen3.6-35B-A3B-MTP-GGUF:UD-Q4_K_SFiles and context on RTX 4090
Each file of Qwen3.6-35B-A3B-MTP-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 4090 | Context (16-bit) | Context (8-bit) |
|---|---|---|---|---|
| BF16 | 71.1 GB | Doesn't fit | — | — |
| UD-Q8_K_XL | 39.1 GB | Doesn't fit | — | — |
| Q8_0 | 37.8 GB | Doesn't fit | — | — |
| UD-Q6_K_XL | 32.6 GB | Doesn't fit | — | — |
| UD-Q6_K | 30.0 GB | Doesn't fit | — | — |
| UD-Q5_K_XL | 27.2 GB | Doesn't fit | — | — |
| UD-Q5_K_M | 27.1 GB | Doesn't fit | — | — |
| UD-Q5_K_S | 25.5 GB | Doesn't fit | — | — |
| UD-Q4_K_XL | 22.9 GB | Doesn't fit | — | — |
| UD-Q4_K_M | 22.7 GB | Doesn't fit | — | — |
| MXFP4_MOE | 22.2 GB | Doesn't fit | — | — |
| UD-Q4_K_S | 21.4 GB | Tight | 7K | 14K |
| UD-IQ4_NL | 18.5 GB | Tight | 143K | 256K |
| UD-IQ4_XS | 18.2 GB | Tight | 159K | 256K |
| UD-Q3_K_XL | 17.2 GB | Fits | 205K | 256K |
| UD-Q3_K_M | 17.1 GB | Fits | 211K | 256K |
| UD-IQ3_S | 15.3 GB | Fits | 256K | 256K |
| UD-IQ3_XXS | 14.1 GB | Fits | 256K | 256K |
| UD-Q2_K_XL | 12.6 GB | Fits | 256K | 256K |
| UD-IQ2_M | 11.9 GB | Fits | 256K | 256K |
| UD-IQ2_XXS | 11.8 GB | Fits | 256K | 256K |
| UD-IQ1_M | 11.4 GB | Fits | 256K | 256K |