cstr/mimo-audio-tokenizer-GGUF overview
MiMo Audio Tokenizer — GGUF GGUF conversion of XiaomiMiMo/MiMo Audio Tokenizer https://huggingface.co/XiaomiMiMo/MiMo Audio Tokenizer encoder only for CrispASR…
Runs locally from ~1.21 GB disk (4 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| mimo-tokenizer.gguf | GGUF | GGUF | 1.21 GB | Download |
Model Details
Model README
---
license: apache-2.0
tags:
- gguf
- audio-tokenizer
- speech-encoder
- mimo
- crispasr
base_model: XiaomiMiMo/MiMo-Audio-Tokenizer
---
MiMo Audio Tokenizer — GGUF
GGUF conversion of XiaomiMiMo/MiMo-Audio-Tokenizer (encoder-only) for CrispASR.
This is a companion model used by the mimo-asr backend — it provides the audio encoder (Whisper-style) that feeds into the MiMo-V2.5-ASR LLM. Not intended for standalone use.
Files
| File | Size | Description |
|------|------|-------------|
| mimo-tokenizer.gguf | 1.2 GB | Audio tokenizer (encoder-only, F16) |
Usage
The tokenizer is loaded automatically by the mimo-asr backend:
crispasr --backend mimo-asr \
-m mimo-v2.5-asr.gguf \
--codec-model mimo-tokenizer.gguf \
-f audio.wavLicense
Apache 2.0 (same as the original model).
Credits
- Original model: XiaomiMiMo
- GGUF conversion & runtime: CrispASR
Run cstr/mimo-audio-tokenizer-GGUF with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models