Update README.md
Browse filesAdd `Quantized GGUF Models` section
README.md
CHANGED
|
@@ -52,3 +52,20 @@ tags:
|
|
| 52 |
wasmedge --dir .:. --nn-preload default:GGML:AUTO:openchat-3.5-0106-Q5_K_M.gguf llama-chat.wasm -p openchat -r '<|end_of_turn|>'
|
| 53 |
```
|
| 54 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 52 |
wasmedge --dir .:. --nn-preload default:GGML:AUTO:openchat-3.5-0106-Q5_K_M.gguf llama-chat.wasm -p openchat -r '<|end_of_turn|>'
|
| 53 |
```
|
| 54 |
|
| 55 |
+
## Quantized GGUF Models
|
| 56 |
+
|
| 57 |
+
| Name | Quant method | Bits | Size | Max RAM required | Use case |
|
| 58 |
+
| ---- | ---- | ---- | ---- | ---- | ----- |
|
| 59 |
+
| [openchat-3.5-0106.Q2_K.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q2_K.gguf) | Q2_K | 2 | 3.08 GB| 5.58 GB | smallest, significant quality loss - not recommended for most purposes |
|
| 60 |
+
| [openchat-3.5-0106.Q3_K_L.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q3_K_L.gguf) | Q3_K_L | 3 | 3.82 GB| 6.32 GB | small, substantial quality loss |
|
| 61 |
+
| [openchat-3.5-0106.Q3_K_M.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q3_K_M.gguf) | Q3_K_M | 3 | 3.52 GB| 6.02 GB | very small, high quality loss |
|
| 62 |
+
| [openchat-3.5-0106.Q3_K_S.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q3_K_S.gguf) | Q3_K_S | 3 | 3.16 GB| 5.66 GB | very small, high quality loss |
|
| 63 |
+
| [openchat-3.5-0106.Q4_0.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q4_0.gguf) | Q4_0 | 4 | 4.11 GB| 6.61 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
|
| 64 |
+
| [openchat-3.5-0106.Q4_K_M.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q4_K_M.gguf) | Q4_K_M | 4 | 4.37 GB| 6.87 GB | medium, balanced quality - recommended |
|
| 65 |
+
| [openchat-3.5-0106.Q4_K_S.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q4_K_S.gguf) | Q4_K_S | 4 | 4.14 GB| 6.64 GB | small, greater quality loss |
|
| 66 |
+
| [openchat-3.5-0106.Q5_0.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q5_0.gguf) | Q5_0 | 5 | 5.00 GB| 7.50 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
|
| 67 |
+
| [openchat-3.5-0106.Q5_K_M.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q5_K_M.gguf) | Q5_K_M | 5 | 5.13 GB| 7.63 GB | large, very low quality loss - recommended |
|
| 68 |
+
| [openchat-3.5-0106.Q5_K_S.gguf](https://huggingface.co/second-state/OpenChat-3.5-0106-GGUF/blob/main/openchat-3.5-0106-Q5_K_S.gguf) | Q5_K_S | 5 | 5.00 GB| 7.50 GB | large, low quality loss - recommended |
|
| 69 |
+
|
| 70 |
+
**Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
|
| 71 |
+
|