ggml-org/MiMo-V2.6-Flash-RL-GGUF

MiMo-V2.6-Flash-RL

Run with https://llama.app

llama serve -hf ggml-org/MiMo-V2.6-Flash-RL-GGUF

Source models

• —https://huggingface.co/XiaomiMiMo/MiMo-V2.6-Flash-RL

Notes

• —The MXFP4 output keeps the routed experts at their native MXFP4 precision.
• —The Q2K output keeps the expert down projections at MXFP4, and quantizes the gate/up projections to Q2K.
• —Includes MTP sidecars (MXFP4 and Q8_0) for speculative decoding (--mtp).
• —Includes a Q8_0 mmproj for the vision and audio encoders.
• —Currently, the Q2 models do not use an imatrix calibration due to lack of one.

[!IMPORTANT] This model is automatically converted using https://github.com/ggml-org/convert