tozp/sensevoice-small-onnx-int8-qdq
automatic-speech-recognition model by tozp
Older snapshot: check the original source for current information.
From the publisher
Model card & documentation
Source previewRead the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.
This repository provides an Explicit INT8 QDQ (QuantizeLinear / DequantizeLinear) static quantization of FunAudioLLM/SenseVoiceSmall. Unlike typical CPU-only dynamic INT8 models (which rely on MatMulInteger and cause subgraph fragmentation and out-of-memory errors on GPU compilers), this model adheres strictly to the NVIDIA TensorRT Explicit QDQ specification. It achieves hardware-level Tensor Core fusion on NVIDIA GPUs while maintaining 100% native CPU compatibility through…
Read the full model card ↗ · Preview checked 2026-11-03T22:19:19.392Z
Inside the original model card — Document outline
- SenseVoiceSmall INT8 QDQ (TensorRT & Universal CPU)
- Key Highlights
- Benchmark & Fidelity Evaluation
- 1. Accuracy & Numerical Fidelity
- 2. Model Size Comparison
- Graph Contract
- Quick Start & Usage
- 1. Running on CPU (Universal standard computer)
- 2. Running on NVIDIA GPU via TensorRT / CUDA
- Citation
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Documentation belongs to its respective authors. Reported project/model license: other. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
tozp/sensevoice-small-onnx-int8-qdq is a automatic-speech-recognition model repository published by tozp on Hugging Face. The source reports the tensorrt library.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- automatic-speech-recognition
- Library
- tensorrt
- Recent downloads (30 days)
- 0
- Cumulative likes
- 0
- Hugging Face trending score
- 0
- Reported safetensors parameters
- Not reported
- Architecture
- Not reported
- License
- other
- Access
- Not gated by Hugging Face
- Created
- 2026-09-22T06:13:10.000Z
- Last modified
- 2026-09-22T06:15:26.000Z
Compatibility and lineage
Reported languages: zh, en, yue, ja, ko
Reported base models: FunAudioLLM/SenseVoiceSmall
Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.
Use and evaluate
Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.
No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.
Source and freshness
Source: Hugging Face Hub. Metadata observed 2026-09-22T06:37:17.952Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.
Related models
jonatasgrosman/wav2vec2-large-xlsr-53-japanese