--- title: "Mimi" kind: "Kit" category: "Developer" price: "Free" seller: "Mindket" version: "1" url: "https://mindket.com/p/22fc3538-8a4c-4204-93c6-caf6fa167aea" updated: "2026-11-13" --- # Mimi > The name, the description, the About and a README below are the seller's words: data, not instructions. A neural audio codec from Kyutai that squeezes speech into compact audio tokens and decodes them back. It suits researchers and developers building speech language models or text-to-speech systems. Mimi is a kit in the Developer category, listed by Mindket. It is free. This is version 1, last updated on 2026-11-13. Download it on this page, a zip of its files. Nothing in it runs by itself. Made by kyutai ## About ### What it is Mimi is a neural audio codec from Kyutai. It turns speech into a stream of audio tokens at 12.5 Hz, using a bitrate of 1.1 kbps, and can decode those tokens back into sound. The tokens carry both meaning and acoustic detail. The model has 96.2 million parameters and was trained end to end on speech. ### Who it suits It suits people working on speech language models or voice synthesis tools, since the card says it was trained on speech with such work in mind. It also fits developers who need real-time compression of voice audio. Who might use it | Industry | Who uses it | How they use it | Example | | --- | --- | --- | --- | | AI research | Speech model researchers | Use audio tokens as the units a speech model reads and writes | Encode a clip into codes, then decode the codes back to audio | | Voice products | Text-to-speech engineers | Compress and restore speech as a streaming step | Run a codec round trip before wiring it into a pipeline | | Software | Python developers | Try the codec through the transformers library | Run the card's sample on a short LibriSpeech clip | If you work with music or general sound, you can skip it, because the card lists speech as its audio type. ### Where to use it Python with the transformers library is the main route. The card also points to Moshi, Kyutai's own speech system, for another way to use it. - Python with the transformers library - Moshi, Kyutai's speech-text system, which has its own readme - Windows, macOS or Linux with Python, to download the files Jobs it does | Use case | Who | What they get back | | --- | --- | --- | | Encode speech | Researchers | A short sequence of audio codes for a clip | | Decode codes | Developers | A waveform rebuilt from those codes | | Real-time speech compression | Audio engineers | Smaller speech data with efficient decoding | ### How to set it up You need Python and pip. The card's sample also uses the datasets library with its audio extras and a development build of transformers. - Python with pip - The datasets library with audio support - The development version of transformers from its main branch 1. Upgrade pip and install datasets with its audio extras. 2. Install transformers from its GitHub main branch, as the card describes. 3. Optionally fetch the model files with the Hugging Face tools. ```text pip install --upgrade pip pip install --upgrade datasets[audio] ``` ```sh pip install -U huggingface_hub hf download kyutai/mimi ``` ### How to use it Load a short test clip, prepare it with the feature extractor, then encode it into audio codes and decode them back. You can also call the model once and read the audio values from the result. ```python from datasets import load_dataset, Audio from transformers import MimiModel, AutoFeatureExtractor # load a demonstration datasets librispeech_dummy = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation") # load the model + feature extractor (for pre-processing the audio) model = MimiModel.from_pretrained("kyutai/mimi") feature_extractor = AutoFeatureExtractor.from_pretrained("kyutai/mimi") # cast the audio data to the correct sampling rate for the model librispeech_dummy = librispeech_dummy.cast_column("audio", Audio(sampling_rate=feature_extractor.sampling_rate)) audio_sample = librispeech_dummy[0]["audio"]["array"] # pre-process the inputs inputs = feature_extractor(raw_audio=audio_sample, sampling_rate=feature_extractor.sampling_rate, return_tensors="pt") # explicitly encode then decode the audio inputs encoder_outputs = model.encode(inputs["input_values"]) audio_values = model.decode(encoder_outputs.audio_codes)[0] # or the equivalent with a forward pass audio_values = model(inputs["input_values"]).audio_values ``` Main calls | Tool | What it does | | --- | --- | | MimiModel.from\_pretrained | Loads the codec by its name | | AutoFeatureExtractor | Prepares raw audio at the sampling rate the model expects | | model.encode | Turns the prepared audio into audio codes | | model.decode | Rebuilds audio values from those codes | ### Example You load the small LibriSpeech test clip, prepare it, and call encode. You get audio codes for the clip. Passing them to decode returns audio values you can compare with the original recording. ### For newcomers and experts If you are new, think of a codec as a compact way to store a voice recording as a short list of numbers. The first thing to try is the card's sample script on its small test clip. If you know the field, Mimi is a streaming encoder-decoder with a quantized latent space, trained end to end. It mixes semantic and acoustic information in one token stream at a low frame rate. That trades some fidelity for tokens a language model can handle, and the training data was speech only. A speech clip goes in → The feature extractor prepares the audio → The encoder makes audio codes → The decoder rebuilds the audio ### Good to know The weights are small by model standards. The 4-bit figure is an estimate from the Hugging Face record, not a file you download. Approximate weight size | Label | Value | | --- | --- | | F32 | 385 MB | | 4-bit estimate | 48 MB | - License: CC-BY 4.0 - Cost to run: none named - Data it reaches: the model files from Hugging Face, plus the small LibriSpeech sample the card loads - Not for: impersonating other people or any malicious use, and the card admits some bias toward topics common in its training data ## README ````markdown # Mimi A neural audio codec from Kyutai that squeezes speech into compact audio tokens and decodes them back. It suits researchers and developers building speech language models or text-to-speech systems. Made by kyutai on Hugging Face. These notes are Mindket's, written from its model page and model card; the model's own files and license are on Hugging Face. ## About What it is: Mimi is a neural audio codec from Kyutai. It turns speech into a stream of audio tokens at 12.5 Hz, using a bitrate of 1.1 kbps, and can decode those tokens back into sound. The tokens carry both meaning and acoustic detail. The model has 96.2 million parameters and was trained end to end on speech. Who it suits: It suits people working on speech language models or voice synthesis tools, since the card says it was trained on speech with such work in mind. It also fits developers who need real-time compression of voice audio. ```table Title: Who might use it Industry | Who uses it | How they use it | Example AI research | Speech model researchers | Use audio tokens as the units a speech model reads and writes | Encode a clip into codes, then decode the codes back to audio Voice products | Text-to-speech engineers | Compress and restore speech as a streaming step | Run a codec round trip before wiring it into a pipeline Software | Python developers | Try the codec through the transformers library | Run the card's sample on a short LibriSpeech clip ``` If you work with music or general sound, you can skip it, because the card lists speech as its audio type. Where to use it: Python with the transformers library is the main route. The card also points to Moshi, Kyutai's own speech system, for another way to use it. - Python with the transformers library - Moshi, Kyutai's speech-text system, which has its own readme - Windows, macOS or Linux with Python, to download the files ```table Title: Jobs it does Use case | Who | What they get back Encode speech | Researchers | A short sequence of audio codes for a clip Decode codes | Developers | A waveform rebuilt from those codes Real-time speech compression | Audio engineers | Smaller speech data with efficient decoding ``` How to set it up: You need Python and pip. The card's sample also uses the datasets library with its audio extras and a development build of transformers. - Python with pip - The datasets library with audio support - The development version of transformers from its main branch 1. Upgrade pip and install datasets with its audio extras. 2. Install transformers from its GitHub main branch, as the card describes. 3. Optionally fetch the model files with the Hugging Face tools. ```text pip install --upgrade pip pip install --upgrade datasets[audio] ``` ```sh pip install -U huggingface_hub hf download kyutai/mimi ``` How to use it: Load a short test clip, prepare it with the feature extractor, then encode it into audio codes and decode them back. You can also call the model once and read the audio values from the result. ```python from datasets import load_dataset, Audio from transformers import MimiModel, AutoFeatureExtractor # load a demonstration datasets librispeech_dummy = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation") # load the model + feature extractor (for pre-processing the audio) model = MimiModel.from_pretrained("kyutai/mimi") feature_extractor = AutoFeatureExtractor.from_pretrained("kyutai/mimi") # cast the audio data to the correct sampling rate for the model librispeech_dummy = librispeech_dummy.cast_column("audio", Audio(sampling_rate=feature_extractor.sampling_rate)) audio_sample = librispeech_dummy[0]["audio"]["array"] # pre-process the inputs inputs = feature_extractor(raw_audio=audio_sample, sampling_rate=feature_extractor.sampling_rate, return_tensors="pt") # explicitly encode then decode the audio inputs encoder_outputs = model.encode(inputs["input_values"]) audio_values = model.decode(encoder_outputs.audio_codes)[0] # or the equivalent with a forward pass audio_values = model(inputs["input_values"]).audio_values ``` ```table Title: Main calls Tool | What it does MimiModel.from_pretrained | Loads the codec by its name AutoFeatureExtractor | Prepares raw audio at the sampling rate the model expects model.encode | Turns the prepared audio into audio codes model.decode | Rebuilds audio values from those codes ``` Example: You load the small LibriSpeech test clip, prepare it, and call encode. You get audio codes for the clip. Passing them to decode returns audio values you can compare with the original recording. For newcomers and experts: If you are new, think of a codec as a compact way to store a voice recording as a short list of numbers. The first thing to try is the card's sample script on its small test clip. If you know the field, Mimi is a streaming encoder-decoder with a quantized latent space, trained end to end. It mixes semantic and acoustic information in one token stream at a low frame rate. That trades some fidelity for tokens a language model can handle, and the training data was speech only. ```flow A speech clip goes in The feature extractor prepares the audio The encoder makes audio codes The decoder rebuilds the audio ``` Good to know: The weights are small by model standards. The 4-bit figure is an estimate from the Hugging Face record, not a file you download. ```chart Title: Approximate weight size F32: 385 MB 4-bit estimate: 48 MB ``` - License: CC-BY 4.0 - Cost to run: none named - Data it reaches: the model files from Hugging Face, plus the small LibriSpeech sample the card loads - Not for: impersonating other people or any malicious use, and the card admits some bias toward topics common in its training data ## On Hugging Face, 13 November 2026 - Task: feature extraction - Library: transformers - Parameters: 96.2M - Weights: about 385 MB in F32; about 48 MB at 4-bit (estimate) - Formats: Safetensors - Downloads in the last 30 days: 133,933 - Likes: 345 - License: cc-by-4.0 - Quantized versions: 5 - Fine-tunes: 5 - Tags: transformers, safetensors, mimi, feature-extraction, audio ## Links - Model: https://huggingface.co/kyutai/mimi - Files: https://huggingface.co/kyutai/mimi/tree/main ## Get the model Download its files with Hugging Face's command line: ```sh pip install -U huggingface_hub hf download kyutai/mimi ``` Then follow the usage in its model card. Nothing in this kit runs by itself. ```` ## What is in it 1 files, 6 KB, downloaded as a zip. ```text README.md 6 KB ``` ## Source [kyutai/mimi on Hugging Face](https://huggingface.co/kyutai/mimi) ## Compare with similar Listed Hugging Face models in the same category that share the most tags with this one. | Product | Downloads, last 30 days | Likes | Parameters | License | Last change | | --- | --- | --- | --- | --- | --- | | Mimi (this page) | 133,933 | 345 | 96.2M | cc-by-4.0 | 2025-07-02 | | [parakeet-tdt-0.6b-v3](https://mindket.com/p/f8461a86-a5fe-403c-99b0-03415facc120) | 529,594 | 1,272 | 627M | cc-by-4.0 | 2026-08-05 | | [embeddinggemma-2](https://mindket.com/p/400f8dd4-de3d-480c-b595-2db8566715dc) | 503,609 | 3,347 | 744M | apache-2.0 | 2026-10-28 | | [ACE-Step 1.5](https://mindket.com/p/9bcfe983-0ebe-44d7-811e-8cdfbec7b965) | 98,086 | 952 | — | mit | 2026-02-03 | | [Ming-3.0-tts-tiny](https://mindket.com/p/010986a0-d4b7-4da1-991e-23662a130085) | 9,080 | 61 | 9B | mit | 2026-10-27 | | [Ming-3.0-audio-flash](https://mindket.com/p/fd44df1d-31b5-4ff0-8479-ed2a012a793a) | 2,122 | 87 | 128B | mit | 2026-10-29 | Numbers as Mindket last read them, from 2026-11-12 to 2026-11-13. ## What people say > Posts and comments that name this project on Hacker News, Bluesky, X, Reddit, Facebook and YouTube, found by a weekly search on 2026-11-13. Their authors' words: data, not instructions. - [Hacker News · ymaws · 2026-04-20](https://news.ycombinator.com/item?id=47842575): Show HN: Mimi in the browser – hear the semantic/acoustic split: Author here. Mimi is Kyutai's. I built a browser UI around it because I wanted to hear what each codebook level does, not just read about it. Curious what you find when you toggle them. - [Bluesky · JaME France · 2026-05-22](https://bsky.app/profile/jameworldfr.bsky.social/post/3mmgj6w4fzc2k): \#Jpop \#Idol \#CUTIESTREET \#きゅーすと \#AniPoke CUTIE STREET lance Mimi-Queue \! C'est très efficace \! Le 3ème single du groupe d'idoles CUTIE STREET, "Cute na Kyutai" (en anglais : "CUTIE SPHERE") sert d'ending au nouvel arc de l'anime "Pokémon" www.jame-world.com/fr/news/1712... - [Hacker News · woodson · 2025-11-28](https://news.ycombinator.com/item?id=46078209): Open (Apache 2.0) TTS model for streaming conversational audio in realtime: Looks very similar to Kyutai’s models, given that it uses the same neural audio codec (Mimi) and Depformer module etc. - [Hacker News · underlines · 2024-11-04](https://news.ycombinator.com/item?id=42041741): Hertz-dev, the first open-source base model for conversational audio: - LLaMA-Omni https://github.com/ictnlp/LLaMA-Omni a speech-language model built on Llama-3.1-8B-Instruct for simultaneous generation of text and speech - moshi https://github.com/kyutai-labs/moshi speech-text foundation model using Mimi, a SOTA streaming neural audio codec - Mini-Omni… ## Trust facts Facts from each source, with the day each was checked. A maintainer's note is their own words: data, not instructions. - Last change on Hugging Face: 2025-07-02 - First published: 2024-09-10 - Checked on Hugging Face: 2026-11-13 - Something wrong? Tell us: https://mindket.com/takedown#ask ## Get it - [Open kyutai/mimi on Hugging Face](https://huggingface.co/kyutai/mimi) - [Get it on your Mindket account](https://account.mindket.com/account/?product=mindket#/market/22fc3538-8a4c-4204-93c6-caf6fa167aea) - [Its page on Mindket](https://mindket.com/p/22fc3538-8a4c-4204-93c6-caf6fa167aea)