Quantized
Collection
A collection of quantized model I found.
•
8 items
•
Updated
This repository contains GGUF weights for ThaiLLM/ThaiLLM-30B.
For a convenient overview and download list, visit our model page.
If you are unsure how to use GGUF files, refer to the llama.cpp documentation for more details.
./llama-cli -m ThaiLLM-30B-q4_k_m.gguf -p "Hello!"
(sorted by size, not necessarily quality)
| Link | Type | Size/GB | Notes |
|---|---|---|---|
| GGUF | q2_k | 10.49 | very low quality, for testing |
| GGUF | q3_k_m | 13.70 | |
| GGUF | q4_0 | 16.12 | |
| GGUF | q4_k_m | 17.28 | recommended, good balance |
| GGUF | q5_k_m | 20.23 | |
| GGUF | q8_0 | 30.25 | near-full precision |
Special thanks to the llama.cpp team for their amazing work.
2-bit
3-bit
4-bit
5-bit
8-bit
Base model
ThaiLLM/ThaiLLM-30B