Instructions to use deagentai/lobe3 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use deagentai/lobe3 with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="deagentai/lobe3",
	filename="ggml-model-Q4_0.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 deagentai/lobe3 with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf deagentai/lobe3:Q4_0
# Run inference directly in the terminal:
llama-cli -hf deagentai/lobe3:Q4_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf deagentai/lobe3:Q4_0
# Run inference directly in the terminal:
llama-cli -hf deagentai/lobe3:Q4_0

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 deagentai/lobe3:Q4_0
# Run inference directly in the terminal:
./llama-cli -hf deagentai/lobe3:Q4_0

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 deagentai/lobe3:Q4_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf deagentai/lobe3:Q4_0

Use Docker

docker model run hf.co/deagentai/lobe3:Q4_0

LM Studio
Jan
Ollama
How to use deagentai/lobe3 with Ollama:
```
ollama run hf.co/deagentai/lobe3:Q4_0
```

Unsloth Studio new

How to use deagentai/lobe3 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 deagentai/lobe3 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 deagentai/lobe3 to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for deagentai/lobe3 to start chatting

Docker Model Runner
How to use deagentai/lobe3 with Docker Model Runner:
```
docker model run hf.co/deagentai/lobe3:Q4_0
```

Lemonade

How to use deagentai/lobe3 with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull deagentai/lobe3:Q4_0

Run and chat with the model

lemonade run user.lobe3-Q4_0

List all available models

lemonade list

deagentai commited on Feb 18, 2025

Commit

0201751

verified ·

1 Parent(s): ea461a3

Update README.md

Browse files

Files changed (1) hide show

README.md +2 -8

README.md CHANGED Viewed

@@ -13,7 +13,8 @@ license: mit
 ## Model Description
-This model is a fine-tuned version of `deepseek-ai/DeepSeek-R1-Distill-Qwen-7B`, specifically optimized for cryptocurrency, Web3, and DeFi-related tasks. It is designed to provide specialized knowledge while maintaining computational efficiency.
 ### Supported Tasks
@@ -29,13 +30,6 @@ This model is a fine-tuned version of `deepseek-ai/DeepSeek-R1-Distill-Qwen-7B`,
 The model is fine-tuned on specialized crypto and web3 prompts. Training data should be placed in the `./json/` directory in JSONL format.
-## Technical Specifications
-- **Base Model**: deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
-- **Training Type**: Supervised Fine-Tuning (SFT)
-- **Required Python Version**: 3.13+
-- **Package Manager**: uv
 ## Performance and Limitations
 The model is optimized for:

 ## Model Description
+This model is a fine-tuned version of `Qwen-7B`, specifically optimized for cryptocurrency, Web3, and DeFi-related tasks.
+It is designed to provide specialized knowledge and decision ability while maintaining computational efficiency.
 ### Supported Tasks
 The model is fine-tuned on specialized crypto and web3 prompts. Training data should be placed in the `./json/` directory in JSONL format.
 ## Performance and Limitations
 The model is optimized for: