|
--- |
|
license: apache-2.0 |
|
language: |
|
- en |
|
pipeline_tag: text-generation |
|
tags: |
|
- llm |
|
- ggml |
|
--- |
|
|
|
|
|
# GGML converted versions of [Mosaic's](https://huggingface.co/mosaicml) MPT Models |
|
|
|
|
|
## CAUTION: MPT Development is still ongoing and not finished! |
|
- Rust & Python: Rustformers implementation see here: [Implement MPT Model](https://github.com/rustformers/llm/pull/218) |
|
|
|
If these implementations are complete i will add instructions on how to run the models and update them if necesary! |
|
|
|
|
|
## Converted Models: |
|
|
|
|
|
| Name | Based on | Type | Container | |
|
|-|-|-|-| |
|
| [mpt-7b-f16.bin](https://huggingface.co/LLukas22/mpt-7b-ggml/blob/main/mpt-7b-f16.bin) | [mpt-7b](https://huggingface.co/mosaicml/mpt-7b) | fp16 | GGML | |
|
| [mpt-7b-q4_0-ggjt.bin](https://huggingface.co/LLukas22/mpt-7b-ggml/blob/main/mpt-7b-q4_0-ggjt.bin) | [mpt-7b](https://huggingface.co/mosaicml/mpt-7b) | int4 | GGJT | |
|
| [mpt-7b-chat-f16.bin](https://huggingface.co/LLukas22/mpt-7b-ggml/blob/main/mpt-7b-chat-f16.bin) | [mpt-7b-chat](https://huggingface.co/mosaicml/mpt-7b-chat) | fp16 | GGML | |
|
| [mpt-7b-chat-q4_0-ggjt.bin](https://huggingface.co/LLukas22/mpt-7b-ggml/blob/main/mpt-7b-chat-q4_0-ggj.bin) | [mpt-7b-chat](https://huggingface.co/mosaicml/mpt-7b-chat) | int4 | GGJT | |
|
| [mpt-7b-instruct-f16.bin](https://huggingface.co/LLukas22/mpt-7b-ggml/blob/main/mpt-7b-instruct-f16.bin) | [mpt-7b-instruct](https://huggingface.co/mosaicml/mpt-7b-instruct) | fp16 | GGML | |
|
| [mpt-7b-instruct-q4_0-ggjt.bin](https://huggingface.co/LLukas22/mpt-7b-ggml/blob/main/mpt-7b-instruct-q4_0-ggjt.bin) | [mpt-7b-instruct](https://huggingface.co/mosaicml/mpt-7b-instruct) | int4 | GGJT | |
|
| [mpt-7b-storywriter-f16.bin](https://huggingface.co/LLukas22/mpt-7b-ggml/blob/main/mpt-7b-f16.bin) | [mpt-7b-storywriter](https://huggingface.co/mosaicml/mpt-7b-storywriter) | fp16 | GGML | |
|
| [mpt-7b-storywriter-q4_0-ggjt.bin](https://huggingface.co/LLukas22/mpt-7b-ggml/blob/main/mpt-7b-storywriter-q4_0-ggjt.bin) | [mpt-7b-storywriter](https://huggingface.co/mosaicml/mpt-7b-storywriter) | int4 | GGJT | |
|
|
|
|
|
## Usage |
|
|
|
### Rust & Python: |
|
#### TBD See above! |
|
|
|
### Via GGML |
|
The `GGML` example only supports the ggml container type! |
|
|
|
##### Installation |
|
|
|
``` |
|
git clone https://github.com/ggerganov/ggml |
|
cd ggml |
|
mkdir build && cd build |
|
cmake .. |
|
make -j4 mpt |
|
``` |
|
|
|
#### Run inference |
|
|
|
``` |
|
./bin/mpt -m path/to/model.bin -p "The meaning of life is" |
|
``` |
