Model reference · open weights
Audio8-ASR is an open-weight audio or speech model from Audio8, listed in the AxForge catalogue. AxForge can bring it up on EU-owned hardware for you on request — with the licence handled where one is required.
About
Audio8-ASR-0.1B [](https://github.com/AutoArk/open-audio-opd) [](https://arxiv.org/abs/2605.28139) [](https://creativecommons.org/licenses/by-nc/4.0/) Audio8-ASR-0.1B is a compact autoregressive ASR model whose language-model component has only 0.1B parameters. It supports multilingual speech recognition for languages including Chinese, English, French, German, Japanese, Korean, and Cantonese. We position it as one of the smallest usable performance ASR models in the LLM era. This base repository provides the Hugging Face Transformers checkpoint. We also provide deployment-focused releases: - Audio8-ASR-0.1B-onnx-runtime - Audio8-ASR-0.1B-iOS-ANE The ONNX Runtime release is designed for edge-device deployment and can run with roughly 1.1 GB peak memory footprint, depending on device, runtime configuration, and workload. The iOS release is designed for local iPhone transcription with roughly 200 MB peak runtime memory footprint, depending on device, iOS version, and workload. Demo Video Your browser does not support the video tag. Evaluation Results The Open ASR results use the seven current public splits from hf-audio/open-asr-leaderboard at dataset revision b6bdcd0beb34f8975dc659796176d88f43aff502. They were measured with the standalone Transformers package on standardized H200 Hugging Face Jobs using BF16, eager attention, greedy decoding, maxnewtokens=256, and the documented 30-second audio cap. Per-split batch sizes were 1152, 1024, 1408, 1024, 1024, 2048, and 628. Raw manifests are stored in hf://buckets/AutoArk-AI/audio8-asr-open-asr-results, and the corresponding machine-readable results are provided in .evalresults/openasrleaderboard.yaml. The internal canonical WenetSpeech results come from the reproducibility-checked teacher0p6B-step3000 export with batch size 128. Its effective model tensors are byte-identical to this standalone release; the release only removes a redundant tied LM-head tensor and packages the same weights for standalone use. Chinese results are reported as character error rate. AISHELL is intentionally excluded from this table. Model Overview - Task: automatic speech recognition - Checkpoint format: safetensors - Sampling rate: 16 kH
Summarised from the published model card. Read the full card on the HuggingFace links below.
Specifications
| Maker | Audio8 |
|---|---|
| Type | Audio & music |
| Parameters (lead) | 324M |
| Variants | 1 |
| Runs with | transformers |
| Released | 2026-07-10 |
| Popularity | 2k downloads / month |
| Likes | 79 |
| Licence | Commercial licence needed |
How it works
Variants
Open weights ship in several sizes and precisions. One page, all the variants — pick the one that fits your GPU. VRAM figures are estimates from model size.
| Variant | Params | Precision | VRAM | Fits 16 GB | Weights |
|---|---|---|---|---|---|
| Audio8-ASR-0.1B | 324M | BF16 | ~0.7 GB | ✓ | Weights ↗ |
Using it via the API
Once AxForge deploys audio8-asr for you, it answers on the OpenAI-compatible API — the same base URL and keys as every other model. (audio8-asr below is illustrative; you get the exact model name on deployment.)
$ curl -sS https://api.axforge.ai/v1/audio/transcriptions \ -H "Authorization: Bearer $AXFORGE_API_KEY" \ -F model="audio8-asr" -F file=@audio.mp3
Licence
The weights are open but cc-by-nc-4.0 needs a commercial agreement for business use. AxForge can arrange that licence and host the model for you — you pay AxForge, we settle with the model’s maker. Ask us for a quote. Read the licence ↗