Ling-3.0-tiny: hardware requirements
Ling-3.0-tiny needs about 6.2 GB of memory at Q4 quantization — a 4.8 GB download plus room for context and overhead. That fits a machine with 16GB RAM.
This is a mixture-of-experts model. Only a fraction of its 7.9B parameters are used for each token, so it runs faster than its size suggests — but all of it still has to fit in memory. The speed is that of a small model; the memory requirement is that of a large one. More on this →
Will it run on your machine?
Download sizes by quantization
Lower quantization means a smaller file and less memory, at some cost to quality. Q4_K_M is the usual starting point. What is this?
| Quant | Download | RAM @ 4K | RAM @ 32K | Source |
|---|---|---|---|---|
| Q4_K_Mrecommended | 4.8 GB | 6.2 GB | 12 GB | measured |
| Q5_K_M | 5.6 GB | 7.0 GB | 13 GB | measured |
| Q6_K | 6.5 GB | 8.0 GB | 14 GB | measured |
| Q8_0 | 8.4 GB | 10 GB | 16 GB | measured |
| BF16 | 16 GB | 18 GB | 24 GB | measured |
“Measured” sizes are read from published GGUF files (bloomer010/Ling-3.0-tiny-GGUF). “Estimated” sizes are derived from the parameter count and are typically within a few percent. The RAM columns add the context cache and runtime overhead to the download size.