This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).
Open-weight model · Feature extraction
tiny-audio-granite-qwen-frozen-3
by Alex Kroman mazesmazes/tiny-audio-granite-qwen-frozen-3
tiny-audio-granite-qwen-frozen-3 is an open-weight model for feature extraction from Alex Kroman. It has 15M parameters. At 16-bit it needs about 0 GB of GPU memory, which fits on 1x MI300X from $1.85 an hour, at the lowest prices in the SAVRN Index.
This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model.
Parameters15M
Context—
Weights578.3 MB
License—
AccessOpen weights
Monthly Downloads—
Runs On
What it takes to serve tiny-audio-granite-qwen-frozen-3 (15M 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 (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 0.0 GB | 0.0 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 0.0 GB | 0.0 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x 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, read Oct 12, 2026.
tiny-audio-granite-qwen-frozen-3 on every accelerator the SAVRN Index prices, at every precision
Model Card
This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).
Excerpt from the card by Alex Kroman.
Configuration
- Architecture
- ASRModel
- Vocabulary size
- 248,192
- Model type
- asr_model
Identity and Version
- Repository
- mazesmazes/tiny-audio-granite-qwen-frozen-3
- Publisher
- Alex Kroman
- Task
- Feature extraction
- Modality
- Text
- Library
- transformers
- Parameters
- 15M parameters
- Languages
- Not stated by the source
- Revision
- dd38ad285f0ba53f5db40b57864bcbc6fe9e951e
- First published
- 2026-09-24
- Last updated
- 2026-09-26
Files and Weights
19 files, 598.5 MB in total. The weights are 3 files totalling 578.3 MB in bin, safetensors.
Weights3 files · 578.3 MB
Configuration11 files · 221.2 KB
Tokenizer2 files · 20.0 MB
Documentation1 file · 5.2 KB
Other1 file · 7.7 KB
Repository1 file · 1.6 KB
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| adapter_model.safetensors | Weights | 519.5 MB | 8ce6bde4150a |
| model.safetensors | Weights | 58.8 MB | 16617ea14653 |
| training_args.bin | Weights | 5.3 KB | a647200aa512 |
| adapter_config.json | Configuration | 1.3 KB | — |
| alignment.py | Configuration | 9.4 KB | — |
| asr_config.py | Configuration | 21.0 KB | — |
| asr_modeling.py | Configuration | 120.2 KB | — |
| asr_pipeline.py | Configuration | 17.5 KB | — |
| asr_processing.py | Configuration | 9.0 KB | — |
| config.json | Configuration | 9.1 KB | — |
| diarization.py | Configuration | 25.4 KB | — |
| generation_config.json | Configuration | 231 B | — |
| preprocessor_config.json | Configuration | 457 B | — |
| projectors.py | Configuration | 7.6 KB | — |
| README.md | Documentation | 5.2 KB | — |
| chat_template.jinja | Other | 7.7 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
| tokenizer.json | Tokenizer | 20.0 MB | a454aca253aa |
| tokenizer_config.json | Tokenizer | 1.2 KB | — |
License and Download
- License
- Not stated by the source
- Access
- Open weights, no gate
- Download size
- 578.3 MB
Download from Alex Kroman
Released by Alex Kroman through its official repository on Hugging Face.
Built From
- Described by arXiv:1910.09700
Memory Requirements
| Precision | Weights in memory |
|---|---|
| As published | 578.3 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 tiny-audio-granite-qwen-frozen-3
How much GPU memory does tiny-audio-granite-qwen-frozen-3 need?
About 0 GB at 16-bit and 0 GB at 4-bit: the weights (15M parameters) plus a working margin. A long context needs more.
What is the cheapest GPU to run tiny-audio-granite-qwen-frozen-3 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.
Similar Models
This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).
Standalone ECAPA-TDNN voice encoder extracted from Qwen/Qwen3-TTS-12Hz-1.7B-Base. Produces 2048-dimensional x-vector speaker embeddings from audio. The encoder follows the ECAPA-TDNN architecture (Emphasized Channel Attention, Propagation and Aggregation in TDNN Based Speaker Verification) and uses Res2Net blocks, squeeze-excitation attention, and attentive statistical pooling. Speaker embeddings can be stored and shared as SafeTensors files. These embeddings are designed to drive voice cloning in the Qwen3-TTS family. There are two main inference paths: the qwentts Python package and the vLLM-Omni serving API. The qwentts package wraps the TTS model and exposes generatevoiceclone. To…
More details please refer to our Github: FlagEmbedding. FlagEmbedding can map any text to a low-dimensional dense vector which can be used for tasks like retrieval, classification, clustering, or semantic search. And it also can be used in vector databases for LLMs. Updates - 10/12/2023: Release LLM-Embedder, a unified embedding model to support diverse retrieval augmentation needs for LLMs. Paper:fire: - 09/15/2023: The technical report of BGE has been released - 09/15/2023: The masive training data of BGE has been released - 09/12/2023: New models: - 09/07/2023: Update fine-tune code: Add script to mine hard negatives and support adding instruction during fine-tuning. - 08/09/2023: BGE…
This is the model card of a transformers model that has been pushed on the Hub. Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. Use the code below to get started with the model. Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).
A vision encoder with a runtime-selectable token budget, built to sit between an image and an LLM. A coarse 32px grid covers the whole image; a router spends extra 16px detail only on the regions that need it. Output length is 49 + 3k tokens — 49 at k=0, 121 at k=24 — so you choose the tokens/fidelity trade-off per image, at inference, with no retraining. 24.94M parameters. Trained in 6000 steps (~30 min) on 2× Kaggle T4 by distillation from google/siglip-base-patch16-224 (86M). This is a working, honest baseline, not a competitive encoder. On the Flickr30k official test split it reaches 8.95 mean R@1 against the teacher's 83.25 — roughly 10% of its teacher. Chance on 1000 images is 0.10.…