Instructions to use mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF 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 mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF:Q4_K_M # Run inference directly in the terminal: llama cli -hf mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF:Q4_K_M
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF:Q4_K_M # Run inference directly in the terminal: llama cli -hf mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-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 mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF:Q4_K_M # Run inference directly in the terminal: ./llama-cli -hf mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-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 mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF:Q4_K_M # Run inference directly in the terminal: ./build/bin/llama-cli -hf mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF:Q4_K_M
Use Docker
docker model run hf.co/mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF:Q4_K_M
- LM Studio
- Jan
- Ollama
How to use mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF with Ollama:
ollama run hf.co/mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF:Q4_K_M
- Unsloth Desktop
- Docker Model Runner
How to use mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF with Docker Model Runner:
docker model run hf.co/mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF:Q4_K_M
- Lemonade
How to use mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull mradermacher/oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF:Q4_K_M
Run and chat with the model
lemonade run user.oh-dcft-v3.1-claude-3-5-sonnet-20241022-GGUF-Q4_K_M
List all available models
lemonade list
- Atomic Chat
Is there a safetensors format?
gguf is the quantized version, and I can't do sft training. Even with the conversion, there is some loss, especially in the parameter format. Is there the original safetensors format?
mlfoundations-dev/oh-dcft-v3.1-claude-3-5-sonnet-20241022 this is the original safetensors?
gguf is not a quantized format, it is a container format pretty much like safetensors, just with more features. both safetensors and gguf tensors can be quantized.
as for your question, could you explain what is confusing about the model page? it answers your question in the first sentence.
In fact, I have tried several gguf format models before, but the conversion was not as expected because many gguf-specific quantization types. So I want to find the safetensors format.
I actually want to confirm it. But I don't know why you put the safetensors format in mlfoundations-dev instead of mlfoundations until I read https://huggingface.co/mradermacher/model_requests
I think I understand the problem now: we (mradermacher) only provide gguf quants of other people's models. we didn't put anything on mlfoundations(-dev).