Instructions to use dphn/dolphin-2.9-llama3-8b-gguf with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use dphn/dolphin-2.9-llama3-8b-gguf with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="dphn/dolphin-2.9-llama3-8b-gguf",
	filename="dolphin-2.9-llama3-8b-q3_K_M.gguf",
)

llm.create_chat_completion(
	messages = "No input example has been defined for this model task."
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use dphn/dolphin-2.9-llama3-8b-gguf with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf dphn/dolphin-2.9-llama3-8b-gguf:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf dphn/dolphin-2.9-llama3-8b-gguf:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf dphn/dolphin-2.9-llama3-8b-gguf:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf dphn/dolphin-2.9-llama3-8b-gguf:Q4_K_M

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 dphn/dolphin-2.9-llama3-8b-gguf:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf dphn/dolphin-2.9-llama3-8b-gguf:Q4_K_M

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 dphn/dolphin-2.9-llama3-8b-gguf:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf dphn/dolphin-2.9-llama3-8b-gguf:Q4_K_M

Use Docker

docker model run hf.co/dphn/dolphin-2.9-llama3-8b-gguf:Q4_K_M

LM Studio
Jan
Ollama
How to use dphn/dolphin-2.9-llama3-8b-gguf with Ollama:
```
ollama run hf.co/dphn/dolphin-2.9-llama3-8b-gguf:Q4_K_M
```

Unsloth Studio new

How to use dphn/dolphin-2.9-llama3-8b-gguf 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 dphn/dolphin-2.9-llama3-8b-gguf 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 dphn/dolphin-2.9-llama3-8b-gguf to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for dphn/dolphin-2.9-llama3-8b-gguf to start chatting

Docker Model Runner
How to use dphn/dolphin-2.9-llama3-8b-gguf with Docker Model Runner:
```
docker model run hf.co/dphn/dolphin-2.9-llama3-8b-gguf:Q4_K_M
```

Lemonade

How to use dphn/dolphin-2.9-llama3-8b-gguf with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull dphn/dolphin-2.9-llama3-8b-gguf:Q4_K_M

Run and chat with the model

lemonade run user.dolphin-2.9-llama3-8b-gguf-Q4_K_M

List all available models

lemonade list

Has anyone tried this gguf with agentic framework?

by BigDeeper - opened Apr 30, 2024

Discussion

BigDeeper

Apr 30, 2024

Pythogora (aka gpt-pilot) or Devika?

Crystalcareai

May 1, 2024

If someone were to, I recommend they use the 256k version that we recently posted. It has function calling and agentic abilities in line with: https://huggingface.co/datasets/internlm/Agent-FLAN. Should work pretty well, especially with some good prompting.

BigDeeper

May 1, 2024

This comment has been hidden

BigDeeper

May 1, 2024

If someone were to, I recommend they use the 256k version that we recently posted. It has function calling and agentic abilities in line with: https://huggingface.co/datasets/internlm/Agent-FLAN. Should work pretty well, especially with some good prompting.

I have just tried it, the 256k version, after converting it to Q8_0 version. Unfortunately it does not appear capable of operating stably within gpt-pilot. It gets "confused," promotes looping behavior. It seems to operate normally using ollama cli though. Llama-3-70b-Instruct.Q5_M's operation is much more stable, but it is 50GB and is very slow.

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