How to use from
llama.cppInstall from WinGet (Windows)
winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf ioxdue/moediet-runs:Q4_K_M# Run inference directly in the terminal:
llama cli -hf ioxdue/moediet-runs:Q4_K_MBuild 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 ioxdue/moediet-runs:Q4_K_M# Run inference directly in the terminal:
./build/bin/llama cli -hf ioxdue/moediet-runs:Q4_K_MUse Docker
docker model run hf.co/ioxdue/moediet-runs:Q4_K_MQuick Links
MoEDiet run artifacts
Run outputs of MoEDiet experiments: audited
expert-pruning checkpoints, quantized GGUF exports, masks, and build reports, organized
per job under runs/<JOB_ID>/ (source snapshots of each submitted job are kept under
requests/<ID>/).
Weights are derivatives of Apache-2.0 models (AllenAI OLMoE-1B-7B-0125-Instruct,
Qwen/Qwen3-Next-80B-A3B-Instruct) and are distributed under the same Apache-2.0 terms,
with attribution to the original authors. Each run directory carries its own
report.json with the exact source revision, routing, and quality audit.
- Downloads last month
- 33
Hardware compatibility
Log In to add your hardware
4-bit
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Install (macOS, Linux)
# Start a local OpenAI-compatible server with a web UI: llama serve -hf ioxdue/moediet-runs:Q4_K_M# Run inference directly in the terminal: llama cli -hf ioxdue/moediet-runs:Q4_K_M