# ced-mini by Speech Team, Xiaomi MiLM Plus: Open-Weight Model Source: https://savrn.com/models/ced-mini Markdown alternate of the page above; the site index is https://savrn.com/llms.txt --- ## Runs On What it takes to serve ced-mini (10M parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them. | Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits | | --- | --- | --- | --- | --- | --- | | 16-bit | 0.0 GB | 0.0 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 8-bit | 0.0 GB | 0.0 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | | 4-bit | 0.0 GB | 0.0 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 | Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the [SAVRN Index](https://savrn.com/ai-index/pricing/gpus), read Oct 25, 2026. [ced-mini on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/ced-mini/gpus) ## Model Card By Speech Team, Xiaomi MiLM Plus, published under apache-2.0, revision 26c3ebcae85d. ### CED-Mini Model CED are simple ViT-Transformer-based models for audio tagging, achieving sota performance on Audioset. | Model | Parameters (M) | AS-20K (mAP) | AS-2M (mAP) | | --- | --- | --- | --- | | CED-Tiny | 5.5 | 36.5 | 48.1 | | CED-Mini | 9.6 | 38.5 | 49.0 | | CED-Small | 22 | 41.6 | 49.6 | | CED-Base | 86 | 44.0 | 50.0 | [Read the full model card (325 words)](https://savrn.com/models/ced-mini/card) ## Configuration Architecture CedForAudioClassification Stored precision float32 Model type ced ## Identity and Version Repository mispeech/ced-mini Publisher Speech Team, Xiaomi MiLM Plus Task Audio classification Modality Audio Library transformers Parameters 10M parameters Languages ced Revision 26c3ebcae85d4330f4fc26763f029539a3afcda0 First published 2023-11-01 Last updated 2026-03-30 ## Files and Weights 9 files, 49.8 MB in total. The weights are 2 files totalling 49.7 MB in onnx, safetensors. Weights2 files · 49.7 MB Configuration5 files · 58.7 KB Documentation1 file · 3.4 KB Repository1 file · 1.5 KB Every file | File | Type | Size | SHA-256 | | --- | --- | --- | --- | | model.onnx | Weights | 10.9 MB | 2a86374dbe1f | | model.safetensors | Weights | 38.8 MB | e4070d02ca53 | | config.json | Configuration | 25.1 KB | — | | configuration_ced.py | Configuration | 6.7 KB | — | | feature_extraction_ced.py | Configuration | 6.5 KB | — | | modeling_ced.py | Configuration | 20.1 KB | — | | preprocessor_config.json | Configuration | 384 B | — | | README.md | Documentation | 3.4 KB | — | | .gitattributes | Repository | 1.5 KB | — | ## License and Download License apache-2.0 Access Open weights, no gate Download size 49.7 MB [Download from Speech Team, Xiaomi MiLM Plus](https://huggingface.co/mispeech/ced-mini) Released by Speech Team, Xiaomi MiLM Plus through its official repository on Hugging Face. [Read the license](https://www.apache.org/licenses/LICENSE-2.0). ## Built From - Described by [arXiv:2308.11957](https://savrn.com/papers/ced-consistent-ensemble-distillation-for-audio-tagging) - Trained on (disclosed) AudioSet ## Memory Requirements | Precision | Weights in memory | | --- | --- | | As published | 49.7 MB | | 16-bit | 0.0 GB | | 8-bit | 0.0 GB | | 4-bit | 0.0 GB | Weights only, from the published parameter count; the key-value cache and runtime add to this. ## Questions About ced-mini ### How much GPU memory does ced-mini need? About 0 GB at 16-bit and 0 GB at 4-bit: the weights (10M parameters) plus a working margin. A long context needs more. ### What is the cheapest GPU to run ced-mini on? At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists. ### Can I use ced-mini commercially? Yes. ced-mini is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors. ## Similar Models Model · Audio classification ### [WeSpeaker-ResNet34-LM-MLX](https://savrn.com/models/wespeaker-resnet34-lm-mlx) [Ivan](https://savrn.com/model-publishers/aufklarer) MLX-compatible weights for WeSpeaker ResNet34-LM, converted from the pyannote speaker embedding model with BatchNorm fused into Conv2d. WeSpeaker ResNet34-LM is a speaker embedding model (~6.6M params) that produces 256-dimensional L2-normalized speaker embeddings from audio. Trained on VoxCeleb for speaker verification and diarization. BatchNorm is fused into Conv2d at conversion time — no BN layers in the MLX model. Part of speech-swift. Converts the original pyannote/wespeaker-voxceleb-resnet34-LM checkpoint using a custom unpickler (no pyannote.audio dependency required). Key transformations: - Fuse BatchNorm into Conv2d: wfused = w × γ/√(σ²+ε), bfused = β − μ×γ/√(σ²+ε) - Transpose… Open weights mit 7M parameters mlx [View model](https://savrn.com/models/wespeaker-resnet34-lm-mlx) Model · Audio classification ### [ls_conformer_base_ge2e](https://savrn.com/models/ls-conformer-base-ge2e) [Quan](https://savrn.com/model-publishers/wq2012) model.safetensors: Primary model weights in SafeTensors format - flaxmodel.msgpack: Native Flax Msgpack serialized parameters - config.json: FlaxSpeakerConfig model & training architecture configuration - preprocessorconfig.json: WhisperFeatureExtractor-compatible frontend configuration - evalresults.json: Verified multi-domain evaluation metrics (1-utt & 3-utt EER, minDCF, ROC-AUC, 95% bootstrap CIs) Open weights apache-2.0 5M parameters flaxspeaker [View model](https://savrn.com/models/ls-conformer-base-ge2e) Model · Audio classification ### [voice-gender-classifier](https://savrn.com/models/voice-gender-classifier) [Huh](https://savrn.com/model-publishers/jaesunghuh) This repo contains the inference code to use pretrained human voice gender classifier. - You could also try Huggingface online demo. First, clone the original github repository and install the packages via pip. For those who need pretrained weights, please download it in here State-of-the-art speaker verification model already produces good representation of the speaker's gender. I used the pretrained ECAPA-TDNN from TaoRuijie's repository, added one linear layer to make two-class classifier, and finetuned the model with the VoxCeleb2 dev set. The model achieved 98.7% accuracy on the VoxCeleb1 identification test split. I would like to note the training dataset I've used for this model… Open weights mit 15M parameters transformers [View model](https://savrn.com/models/voice-gender-classifier) Model · Audio classification ### [ls_ecapa_tdnn_base_ge2e](https://savrn.com/models/ls-ecapa-tdnn-base-ge2e) [Quan](https://savrn.com/model-publishers/wq2012) model.safetensors: Primary model weights in SafeTensors format - flaxmodel.msgpack: Native Flax Msgpack serialized parameters - config.json: FlaxSpeakerConfig model & training architecture configuration - preprocessorconfig.json: WhisperFeatureExtractor-compatible frontend configuration - evalresults.json: Verified multi-domain evaluation metrics (1-utt & 3-utt EER, minDCF, ROC-AUC, 95% bootstrap CIs) Open weights apache-2.0 3M parameters flaxspeaker [View model](https://savrn.com/models/ls-ecapa-tdnn-base-ge2e) Model · Audio classification ### [cn_conformer_small_dr_vectors](https://savrn.com/models/cn-conformer-small-dr-vectors) [Quan](https://savrn.com/model-publishers/wq2012) model.safetensors: Primary model weights in SafeTensors format - flaxmodel.msgpack: Native Flax Msgpack serialized parameters - config.json: FlaxSpeakerConfig model & training architecture configuration - preprocessorconfig.json: WhisperFeatureExtractor-compatible frontend configuration - evalresults.json: Verified multi-domain evaluation metrics (1-utt & 3-utt EER, minDCF, ROC-AUC, 95% bootstrap CIs) Open weights apache-2.0 2M parameters flaxspeaker [View model](https://savrn.com/models/cn-conformer-small-dr-vectors) Model · Audio classification ### [ls_conformer_small_dr_vectors](https://savrn.com/models/ls-conformer-small-dr-vectors) [Quan](https://savrn.com/model-publishers/wq2012) model.safetensors: Primary model weights in SafeTensors format - flaxmodel.msgpack: Native Flax Msgpack serialized parameters - config.json: FlaxSpeakerConfig model & training architecture configuration - preprocessorconfig.json: WhisperFeatureExtractor-compatible frontend configuration - evalresults.json: Verified multi-domain evaluation metrics (1-utt & 3-utt EER, minDCF, ROC-AUC, 95% bootstrap CIs) Open weights apache-2.0 2M parameters flaxspeaker [View model](https://savrn.com/models/ls-conformer-small-dr-vectors) ## Speech Team, Xiaomi MiLM Plus [All models and datasets](https://savrn.com/model-publishers/mispeech) ## Versions - [26c3ebcae85d](https://savrn.com/models/ced-mini/versions/26c3ebcae85d) · current 2026-09-18 ## Explore More - [All audio classification models](https://savrn.com/models/tasks/audio-classification) - [All models under apache-2.0](https://savrn.com/models/licenses/apache-2-0) - [Model comparisons](https://savrn.com/models/comparisons) - [The model directory](https://savrn.com/models) - [Open model prices by host](https://savrn.com/ai-index/pricing/open-models) ## Source - Repository metadata, read 2026-10-21. - [Hugging Face record](https://huggingface.co/mispeech/ced-mini) - [How the hub is built](https://savrn.com/model-hub/methodology)