Update README.md

README.md

@@ -28,13 +28,36 @@ This repo is the result of converting to GGML and quantising.
 ## Provided files
 | Name | Quant method | Bits | Size | RAM required | Use case |
 | ---- | ---- | ---- | ---- | ---- | ----- |
-`mpt7b.ggmlv2.ggmlv2.q4_0.bin` | q4_0 | 4bit | 4.21GB | 7.0GB | 4-bit. |
-`mpt7b.ggmlv2.ggmlv2.q4_1.bin` | q4_0 | 4bit | 4.63GB | 7.5GB | 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
-`mpt7b.ggmlv2.ggmlv2.q5_0.bin` | q5_0 | 5bit | 4.63GB | 7.5GB | 5-bit. Higher accuracy, higher resource usage and slower inference. |
-`mpt7b.ggmlv2.ggmlv2.q5_1.bin` | q5_1 | 5bit | 5.06GB | 7.5GB | 5-bit. Even higher accuracy, and higher resource usage and slower inference. |
-`mpt7b.ggmlv2.ggmlv2.q8_0.bin` | q8_0 | 8bit | 7.58GB | 9.0GB | 8-bit. Almost indistinguishable from float16. Huge resource use and slow. Not recommended for normal use. |
-`mpt7b.ggmlv2.ggmlv2.fp16.bin` | fp16 | 16bit | GB | GB | Full 16-bit. |
+`mpt7b.ggmlv2.q4_0.bin` | q4_0 | 4bit | 4.21GB | 7.0GB | 4-bit. |
+`mpt7b.ggmlv2.q4_1.bin` | q4_0 | 4bit | 4.63GB | 7.5GB | 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
+`mpt7b.ggmlv2.q5_0.bin` | q5_0 | 5bit | 4.63GB | 7.5GB | 5-bit. Higher accuracy, higher resource usage and slower inference. |
+`mpt7b.ggmlv2.q5_1.bin` | q5_1 | 5bit | 5.06GB | 7.5GB | 5-bit. Even higher accuracy, and higher resource usage and slower inference. |
+`mpt7b.ggmlv2.q8_0.bin` | q8_0 | 8bit | 7.58GB | 9.0GB | 8-bit. Almost indistinguishable from float16. Huge resource use and slow. Not recommended for normal use. |
+`mpt7b.ggmlv2.fp16.bin` | fp16 | 16bit | GB | GB | Full 16-bit. |
 
+## Compatibilty
+
+These files are **not** compatible with llama.cpp.
+
+Currently they can be used with:
+* The example `mpt` binary provided with [ggml](https://github.com/ggerganov/ggml)
+* [rustformers' llm](https://github.com/rustformers/llm)
+
+As other options become available I will endeavour to update them here (do let me know in the Community tab if I've missed something!)
+
+## How to build, and an example of using the ggml `mpt` binary (command line only):
+
+```
+git clone https://github.com/ggerganov/ggml
+cd ggml
+mkdir build
+cd build
+cmake ..
+cmake --build . --config Release
+bin/mpt -m /path/to/mpt7b.ggmlv2.ggmlv2.q4_0.bin -t 8 -n 512 -p "Write a story about llamas"
+```
+
+Please see the ggml repo for other build options.
 
 # Original model card: MPT-7B