How to use from
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 runmyai/Ling-3.0-Flash-TokenWing:MXFP4_MOE
# Run inference directly in the terminal:
llama cli -hf runmyai/Ling-3.0-Flash-TokenWing:MXFP4_MOE
Install from WinGet (Windows)
winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf runmyai/Ling-3.0-Flash-TokenWing:MXFP4_MOE
# Run inference directly in the terminal:
llama cli -hf runmyai/Ling-3.0-Flash-TokenWing:MXFP4_MOE
Build from source code
git clone https://github.com/ggml-org/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-app
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama serve -hf runmyai/Ling-3.0-Flash-TokenWing:MXFP4_MOE
# Run inference directly in the terminal:
./build/bin/llama cli -hf runmyai/Ling-3.0-Flash-TokenWing:MXFP4_MOE
Use Docker
docker model run hf.co/runmyai/Ling-3.0-Flash-TokenWing:MXFP4_MOE
Quick Links

Ling 3.0 Flash for TokenWing

Ling 3.0 Flash (by inclusionAI) in the layout TokenWing streams from your SSD, so this 124B mixture-of-experts model runs on ordinary PCs, even with less memory than the model's size.

The easiest way to use it is the Run My AI app: it downloads this model and sets everything up for your PC.

  • Weights: MXFP4 (unchanged from the original quantization); the file's layout is changed so experts can be read straight from the SSD, and TokenWing's data for this model is added to it.
  • License: MIT, as the original model. Credit: inclusionAI.
Downloads last month
75,569
GGUF
Model size
127B params
Architecture
bailingmoe3
Hardware compatibility
Log In to add your hardware

4-bit

Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for runmyai/Ling-3.0-Flash-TokenWing

Quantized
(49)
this model