Can I run glm-4-9b-chat-IMat-GGUF on an RTX 4060?
Yes, just
Its Q5_K_S file (6.69 GB) fits in the 7.20 GB an RTX 4060 can use, with room for ≈ 12K tokens of context.
Full analysis of glm-4-9b-chat-IMat-GGUF
Everything that runs on RTX 4060
Get the Q5_K_S file on Hugging Face
ollama run hf.co/legraphista/glm-4-9b-chat-IMat-GGUF:Q5_K_SFiles and context on RTX 4060
Each file of glm-4-9b-chat-IMat-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 4060 | Context (16-bit) | Context (8-bit) |
|---|---|---|---|---|
| BF16 | 18.8 GB | Doesn't fit | — | — |
| glm-4-9b-chat.FP16 | 18.8 GB | Doesn't fit | — | — |
| Q8_0 | 9.99 GB | Doesn't fit | — | — |
| Q6_K | 8.26 GB | Doesn't fit | — | — |
| Q5_K | 7.14 GB | Doesn't fit | 1K | 3K |
| Q5_K_S | 6.69 GB | Tight | 12K | 24K |
| Q4_K | 6.25 GB | Tight | 23K | 45K |
| Q4_K_S | 5.75 GB | Fits | 34K | 69K |
| IQ4_NL | 5.46 GB | Fits | 42K | 83K |
| Q3_K_L | 5.28 GB | Fits | 46K | 91K |
| IQ4_XS | 5.25 GB | Fits | 46K | 93K |
| Q3_K | 5.06 GB | Fits | 51K | 102K |
| IQ3_M | 4.81 GB | Fits | 57K | 114K |
| IQ3_S | 4.59 GB | Fits | 62K | 125K |
| Q3_K_S | 4.59 GB | Fits | 62K | 125K |
| IQ3_XS | 4.43 GB | Fits | 66K | 128K |
| IQ3_XXS | 4.26 GB | Fits | 70K | 128K |
| Q2_K | 3.99 GB | Fits | 76K | 128K |
| Q2_K_S | 3.96 GB | Fits | 77K | 128K |
| IQ2_M | 3.93 GB | Fits | 78K | 128K |
| IQ2_S | 3.77 GB | Fits | 82K | 128K |
| IQ2_XS | 3.61 GB | Fits | 86K | 128K |
| IQ2_XXS | 3.43 GB | Fits | 90K | 128K |
| IQ1_M | 3.22 GB | Fits | 95K | 128K |
| IQ1_S | 3.10 GB | Fits | 98K | 128K |