# bosonai/higgs-audio-v3-8b-stt-v2

Model from Hugging Face for automatic speech recognition: bosonai/higgs-audio-v3-8b-stt-v2 · unknown

## Facts

- Type: Audio
- Task: automatic-speech-recognition
- Framework: unknown
- License: apache-2.0 — Permissive open license: commercial use, modification, and redistribution are permitted. (https://spdx.org/licenses/Apache-2.0.html)
- Source: https://huggingface.co/bosonai/higgs-audio-v3-8b-stt-v2
- Hosting: hosted (11 files)
- Files: 11 hosted of 25 listed
- Needs its own code to run: Uses custom model code (configuration_higgs_audio.py). Plain llama.cpp, Ollama and LM Studio commands are not offered for this model unless a GGUF file is listed.
- File counts: "listed" is every file in the source repository; only hosted files download from Hugging Bay, and the download plan lists the files it selects for one download, so its count can differ.
- Hosted here: 11 files, 17.8 GB
- Full repo upstream: 17.8 GB
- Recommended file: 17.8 GB (4 files)
- Parameters: 8B
- Download: https://huggingbay.xyz/artifact/hf-model-bosonai-higgs-audio-v3-8b-stt-v2/files (download plan: https://huggingbay.xyz/api/artifacts/hf-model-bosonai-higgs-audio-v3-8b-stt-v2/download-plan)
- Per-file SHA-256: available for 11 of 11 hosted files via https://huggingbay.xyz/api/artifacts/hf-model-bosonai-higgs-audio-v3-8b-stt-v2/files?limit=100
- Run-ready: no (Hugging Bay has not certified a run path for this model; hosted bytes are downloadable and hash-checkable only)
- Peer-assisted fallbacks: 0 verified, 0 pending review
- Downloads: 1953
- Stars: 14
- Card completeness: 96%

## Description

External Hugging Face catalog record for bosonai/higgs-audio-v3-8b-stt-v2. Task: automatic speech recognition. Library: unknown. Languages: en. Base model: bosonai/higgs-audio-v3-8b. Includes 8 model-card evaluation signals. Popularity signal at import: 11,534 downloads and 14 likes.

## Model Card Facts

- Declared license: apache-2.0
- Pipeline: automatic-speech-recognition
- Library: unknown
- Base model: bosonai/higgs-audio-v3-8b
- Task categories: automatic-speech-recognition
- Languages: en
- Model-index entries: higgs-audio-v3-8b-stt-v2
- Evaluation signals: 8 model-card eval results
- File hints: 16 upstream file samples

## Evaluation Signals

- AMI (Meetings test): Test WER = 10.14 (Automatic Speech Recognition)
- Earnings-22: Test WER = 8.73 (Automatic Speech Recognition)
- GigaSpeech: Test WER = 8.47 (Automatic Speech Recognition)
- LibriSpeech (clean): Test WER = 1.25 (Automatic Speech Recognition)
- LibriSpeech (other): Test WER = 2.38 (Automatic Speech Recognition)
- SPGI Speech: Test WER = 3.6 (Automatic Speech Recognition)
- tedlium-v3: Test WER = 3.09 (Automatic Speech Recognition)
- Vox Populi: Test WER = 5.92 (Automatic Speech Recognition)

## File Hints

- .gitattributes (1.5 KB)
- README.md (4.2 KB)
- added_tokens.json (1.1 KB)
- attention.py (8.5 KB)
- common.py (1.1 KB)
- config.json (5.8 KB)
- configuration_higgs_audio.py (9.7 KB)
- cuda_graph_runner.py (5.0 KB)
- custom_modules.py (6.0 KB)
- generation_config.json (147 B)
- higgs_audio_collator.py (36 KB)
- merges.txt (1.6 MB)

## Quality Checklist

- Summary: passed
- Description: passed
- Model card metadata: passed
- License: passed
- Provenance: passed
- File inventory: passed
- Hosted bytes: passed
- Scan result: passed
- Manifest signature: passed

## Safety Notes

- External catalog metadata imported from Hugging Face public API on 2026-08-30T08:25:14.940Z.
- Open license detected as apache-2.0.
- Model-index metadata found for higgs-audio-v3-8b-stt-v2.
- Not a run kit: hosted, hash-checked bytes only; run readiness is not certified for this model yet.

## Actions

- Web: https://huggingbay.xyz/artifact/hf-model-bosonai-higgs-audio-v3-8b-stt-v2
- Files API: https://huggingbay.xyz/api/artifacts/hf-model-bosonai-higgs-audio-v3-8b-stt-v2/files?limit=100
- Metadata: https://huggingbay.xyz/api/artifacts/hf-model-bosonai-higgs-audio-v3-8b-stt-v2/metadata.json
- Distribution: https://huggingbay.xyz/api/artifacts/hf-model-bosonai-higgs-audio-v3-8b-stt-v2/distribution
- Manifest: https://huggingbay.xyz/api/manifests/hf-model-bosonai-higgs-audio-v3-8b-stt-v2
- Download plan (per-file SHA-256 + signed download URLs): https://huggingbay.xyz/api/artifacts/hf-model-bosonai-higgs-audio-v3-8b-stt-v2/download-plan
- Hosted download rows: https://huggingbay.xyz/api/downloadable-files?artifactId=hf-model-bosonai-higgs-audio-v3-8b-stt-v2&limit=100
- Request mirror: not applicable.
- Submit peer fallback for review: POST https://huggingbay.xyz/api/artifacts/hf-model-bosonai-higgs-audio-v3-8b-stt-v2/decentralized-fallbacks