mirror of https://github.com/serge-chat/serge.git synced 2024-01-15 09:32:12 +03:00

Files

Olivier DEBAUCHE 98bdb9f8b0 Add support for Mixtral 8x7B-Dolphin 2.5 and WizardLM (#997 )

* Add support for Mixtral 8x7B - Dolphin 2.5

* Add Support for WizardLM

* Add new LLMs

- Add support for Mixtral 8x7B - Dolphin 2.5
- Add support for WizardLM
- Update README.md

* Remove support of WizardLM 7B -UNCENSORED

* Remove support of WizardLM 7B -UNCENSORED

---------

Co-authored-by: Juan Calderon-Perez <835733+gaby@users.noreply.github.com>

2023-12-19 08:27:10 -05:00

3.6 KiB

Raw Blame History

Serge - LLaMA made easy 🦙

Serge is a chat interface crafted with llama.cpp for running GGUF models. No API keys, entirely self-hosted!

🌐 SvelteKit frontend
💾 Redis for storing chat history & parameters
⚙️ FastAPI + LangChain for the API, wrapping calls to llama.cpp using the python bindings

🎥 Demo:

demo.webm

⚡️ Quick start

🐳 Docker:

docker run -d \
    --name serge \
    -v weights:/usr/src/app/weights \
    -v datadb:/data/db/ \
    -p 8008:8008 \
    ghcr.io/serge-chat/serge:latest

🐙 Docker Compose:

services:
  serge:
    image: ghcr.io/serge-chat/serge:latest
    container_name: serge
    restart: unless-stopped
    ports:
      - 8008:8008
    volumes:
      - weights:/usr/src/app/weights
      - datadb:/data/db/

volumes:
  weights:
  datadb:

Then, just visit http://localhost:8008, You can find the API documentation at http://localhost:8008/api/docs

🖥️ Windows

Ensure you have Docker Desktop installed, WSL2 configured, and enough free RAM to run models.

☁️ Kubernetes

Instructions for setting up Serge on Kubernetes can be found in the wiki.

🧠 Supported Models

Category	Models
Alfred	40B
Code	13B, 33B
CodeLLaMA	7B, 7B-Instruct, 7B-Python, 13B, 13B-Instruct, 13B-Python, 34B, 34B-Instruct, 34B-Python
Falcon	7B, 7B-Instruct, 40B, 40B-Instruct
LLaMA 2	7B, 7B-Chat, 7B-Coder, 13B, 13B-Chat, 70B, 70B-Chat, 70B-OASST
Med42	70B
Medalpaca	13B
Meditron	7B, 7B-Chat, 70B
Mistral	7B-V0.1, 7B-Instruct-v0.2, 7B-OpenOrca
Mixtral	8x7B-v0.1, 8x7B-Dophin-2.5, 8x7B-Instruct-v0.1
Neural-Chat	7B-v3.3
Notus	7B-v1
OpenChat	7B-v3.5-1012
OpenLLaMA	3B-v2, 7B-v2, 13B-v2
Orca 2	7B, 13B
Python Code	13B, 33B
PsyMedRP	13B-v1, 20B-v1
Starling LM	7B-Alpha
Vicuna	7B-v1.5, 13B-v1.5, 33B-v1.3, 33B-Coder
WizardLM	7B-v1.0, 13B-v1.2, 70B-v1.0
Zephyr	3B, 7B-Alpha, 7B-Beta

Additional models can be requested by opening a GitHub issue.

⚠️ Memory Usage

LLaMA will crash if you don't have enough available memory for the model:

💬 Support

Need help? Join our Discord

⭐️ Stargazers

🧾 License

Nathan Sarrazin and Contributors. Serge is free and open-source software licensed under the MIT License and Apache-2.0.

🤝 Contributing

If you discover a bug or have a feature idea, feel free to open an issue or PR.

To run Serge in development mode:

git clone https://github.com/serge-chat/serge.git
cd serge/
docker compose -f docker-compose.dev.yml up --build

3.6 KiB Raw Blame History Unescape Escape