Instructions to use ggml-org/models-moved with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ggml-org/models-moved with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="ggml-org/models-moved",
	filename="bert-bge-small/ggml-model-f16-big-endian.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use ggml-org/models-moved with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf ggml-org/models-moved:F16
# Run inference directly in the terminal:
llama cli -hf ggml-org/models-moved:F16

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf ggml-org/models-moved:F16
# Run inference directly in the terminal:
llama cli -hf ggml-org/models-moved:F16

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf ggml-org/models-moved:F16
# Run inference directly in the terminal:
./llama-cli -hf ggml-org/models-moved:F16

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf ggml-org/models-moved:F16
# Run inference directly in the terminal:
./build/bin/llama-cli -hf ggml-org/models-moved:F16

Use Docker

docker model run hf.co/ggml-org/models-moved:F16

LM Studio
Jan
Ollama
How to use ggml-org/models-moved with Ollama:
```
ollama run hf.co/ggml-org/models-moved:F16
```

Unsloth Studio

How to use ggml-org/models-moved with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for ggml-org/models-moved to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for ggml-org/models-moved to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for ggml-org/models-moved to start chatting

Atomic Chat new
Docker Model Runner
How to use ggml-org/models-moved with Docker Model Runner:
```
docker model run hf.co/ggml-org/models-moved:F16
```

Lemonade

How to use ggml-org/models-moved with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull ggml-org/models-moved:F16

Run and chat with the model

lemonade run user.models-moved-F16

List all available models

lemonade list

How to add new files without cloning the entire repo?

by ggerganov - opened Apr 2, 2024

Discussion

ggerganov

ggml-org org Apr 2, 2024

@phymbert Now that we have Grok-1 in the repo, I'm not sure how to commit new files without having to download extra ~150GB of the grok shards. Is there some workaround?

If not, we might want to move the model to a separate repo, or demonstrate the sharding with something smaller.

julien-c

Apr 3, 2024

if using git: you can clone the repo using GIT_LFS_SKIP_SMUDGE=1 env var (only downloads pointers to large files, not the actual files) and you'll still be able to git commit as usual
alternatives to git:
- use our HTTP API: huggingface-cli upload ...
- or just drag'n'drop files from your browser to upload them (the hashing implem should be efficient, but ~150GB still might be a lot)

julien-c

Apr 3, 2024

That being said, we do recommend (as a convention) splitting different models into different repos

ggerganov

ggml-org org Apr 3, 2024

That being said, we do recommend (as a convention) splitting different models into different repos

Noted, will split the models in different repos accordingly

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment