metadata
license: apache-2.0
language:
- en
pipeline_tag: text-generation
tags:
- llm
- ggml
GGML converted versions of Mosaic's MPT Models
MPT-7B is a decoder-style transformer pretrained from scratch on 1T tokens of English text and code. This model was trained by MosaicML.
MPT-7B is part of the family of MosaicPretrainedTransformer (MPT) models, which use a modified transformer architecture optimized for efficient training and inference.
Converted Models:
| Name | Based on | Type | Container |
|---|---|---|---|
| mpt-7b-f16.bin | mpt-7b | fp16 | GGML |
| mpt-7b-q4_0.bin | mpt-7b | int4 | GGML |
| mpt-7b-q4_0-ggjt.bin | mpt-7b | int4 | GGJT |
| mpt-7b-chat-f16.bin | mpt-7b-chat | fp16 | GGML |
| mpt-7b-chat-q4_0.bin | mpt-7b-chat | int4 | GGML |
| mpt-7b-chat-q4_0-ggjt.bin | mpt-7b-chat | int4 | GGJT |
| mpt-7b-instruct-f16.bin | mpt-7b-instruct | fp16 | GGML |
| mpt-7b-instruct-q4_0.bin | mpt-7b-instruct | int4 | GGML |
| mpt-7b-instruct-q4_0-ggjt.bin | mpt-7b-instruct | int4 | GGJT |
| mpt-7b-storywriter-f16.bin | mpt-7b-storywriter | fp16 | GGML |
| mpt-7b-storywriter-q4_0.bin | mpt-7b-storywriter | int4 | GGML |
| mpt-7b-storywriter-q4_0-ggjt.bin | mpt-7b-storywriter | int4 | GGJT |
⚠️Caution⚠️: mpt-7b-storywriter is still under development!
Usage
Python via llm-rs:
Installation
Via pip: pip install llm-rs
Run inference
from llm_rs import Mpt
from huggingface_hub import hf_hub_download
#Download the model
hf_hub_download(repo_id="LLukas22/mpt-7b-ggml", filename="mpt-7b-q4_0-ggjt.bin", local_dir=".")
#Load the model
model = Mpt("mpt-7b-q4_0-ggjt.bin")
#Generate
print(model.generate("The meaning of life is"))
Rust via Rustformers/llm:
Installation
git clone --recurse-submodules [email protected]:rustformers/llm.git
cargo build --release
Run inference
cargo run --release -- mpt infer -m path/to/model.bin -p "Tell me how cool the Rust programming language is:"
C 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"