Instructions to use wardner/Qwen3-ASR with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- RKLLM
How to use wardner/Qwen3-ASR with RKLLM:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- llama.cpp
How to use wardner/Qwen3-ASR with llama.cpp:
Install (macOS, Linux)
curl -LsSf https://llama.app/install.sh | sh # Start a local OpenAI-compatible server with a web UI: llama serve -hf wardner/Qwen3-ASR # Run inference directly in the terminal: llama cli -hf wardner/Qwen3-ASR
Install from WinGet (Windows)
winget install llama.cpp # Start a local OpenAI-compatible server with a web UI: llama serve -hf wardner/Qwen3-ASR # Run inference directly in the terminal: llama cli -hf wardner/Qwen3-ASR
Build from source code
git clone https://github.com/ggml-org/llama.cpp.git cd llama.cpp cmake -B build cmake --build build -j --target llama-app # Start a local OpenAI-compatible server with a web UI: ./build/bin/llama serve -hf wardner/Qwen3-ASR # Run inference directly in the terminal: ./build/bin/llama cli -hf wardner/Qwen3-ASR
Use Docker
docker model run hf.co/wardner/Qwen3-ASR
- Llama new
- LM Studio
- Jan
- Ollama
How to use wardner/Qwen3-ASR with Ollama:
ollama run hf.co/wardner/Qwen3-ASR
- Unsloth Desktop
- Pi
How to use wardner/Qwen3-ASR with Pi:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf wardner/Qwen3-ASR
Configure the model in Pi
# Install Pi: npm install -g @earendil-works/pi-coding-agent # Add to ~/.pi/agent/models.json: { "providers": { "llama-cpp": { "baseUrl": "http://localhost:9931/v1", "api": "openai-completions", "apiKey": "none", "models": [ { "id": "wardner/Qwen3-ASR" } ] } } }Run Pi
# Start Pi in your project directory: pi
- Docker Model Runner
How to use wardner/Qwen3-ASR with Docker Model Runner:
docker model run hf.co/wardner/Qwen3-ASR
- Lemonade
How to use wardner/Qwen3-ASR with Lemonade:
Pull the model
# Download Lemonade from https://lemonade-server.ai/ lemonade pull wardner/Qwen3-ASR
Run and chat with the model
lemonade run user.Qwen3-ASR-{{QUANT_TAG}}List all available models
lemonade list
- Hermes Agent
How to use wardner/Qwen3-ASR with Hermes Agent:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf wardner/Qwen3-ASR
Configure Hermes
# Install Hermes: curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash hermes setup # Point Hermes at the local server: hermes config set model.provider custom hermes config set model.base_url http://127.0.0.1:9931/v1 hermes config set model.default wardner/Qwen3-ASR
Run Hermes
hermes
- Atomic Chat
- OpenClaw
How to use wardner/Qwen3-ASR with OpenClaw:
Start the llama.cpp server
# Install llama.cpp: brew install llama.cpp # Start a local OpenAI-compatible server: llama serve -hf wardner/Qwen3-ASR
Configure OpenClaw
# Install OpenClaw: npm install -g openclaw@latest # Register the local server and set it as the default model: openclaw onboard --non-interactive --mode local \ --auth-choice custom-api-key \ --custom-base-url http://127.0.0.1:9931/v1 \ --custom-model-id "wardner/Qwen3-ASR" \ --custom-provider-id llama-cpp \ --custom-compatibility openai \ --custom-text-input \ --accept-risk \ --skip-health
Run OpenClaw
openclaw agent --local --agent main --message "Hello from Hugging Face"
Qwen3-ASR — [ASR] Streaming Speech Recognition
1. Overview
Streaming speech recognition: feed in an audio clip, get back the detected language plus the transcript, refreshed once per second (the RK182x streaming deployment of Qwen3-ASR-0.6B).
The upstream Qwen3-ASR family supports 52 languages and dialects (30 languages + 22 Chinese dialects) and handles speech, singing and songs with background music. This repository ships the 0.6B variant running on the streaming path: audio is decoded in 1-second chunks and the committed text grows monotonically.
This repository ships the RK182x deployment artifacts (RKNN graphs +
weights) together with the wdnwebui worker manifest
(qwen3-asr.json).
| Item | Value |
|---|---|
| Model ID | wardner/Qwen3-ASR |
| Display name | Qwen3-ASR |
| Model type | ASR (speech → language + text) |
| Compute target | RK182x |
| Host platform | RK3588 |
| Runtime | RKLLM + RKNN |
| Capabilities | swcap_audio_asr |
| Worker factory | qwen3_asr |
| Memory estimate | ~1600 MB |
| Init time | ~25 s |
| Cards required | 1 |
| Total weight size | 1.05 GB |
2. Files
| Path | Description | Size |
|---|---|---|
encoder_online.1828.rknn |
RKNN graph — streaming audio encoder (RK182x compute card, 1.82 TOPS generation) | 3.0 MB |
encoder_online.weight |
Weights paired with the RKNN graph above | 357.5 MB |
llm.1828.rknn |
RKNN graph — Qwen3-ASR-0.6B LLM | 45.3 MB |
llm.weight |
Weights paired with the RKNN graph above | 369.3 MB |
llm.embed.bin |
Token embedding table, fp16 (vocab 151936 × dim 1024) | 296.8 MB |
llm.tokenizer.gguf |
GPT-2 BPE vocabulary + ByteLevel | 5.7 MB |
mel_128_filters.txt |
Mel filterbank (128 mel × 202 fp32) — must stay next to the weights | 653.3 KB |
qwen3-asr.json |
worker manifest — deliberately plain text | 1847 B |
qwen3-asr.jsonis deliberately not in Git LFS. It is the manual of this distribution unit and must stay plain text so it can be diffed, reviewed and previewed on the Hub.
3. Download
hf download wardner/Qwen3-ASR --local-dir ./Qwen3-ASR
Or with git (requires Git LFS):
git clone https://huggingface.co/wardner/Qwen3-ASR
If huggingface.co is unreachable from your network (DNS interception /
blocked access), point the client at a mirror instead:
HF_ENDPOINT=https://hf-mirror.com hf download wardner/Qwen3-ASR \
--local-dir ./Qwen3-ASR
Note: with git clone the weights are Git LFS pointers until
git lfs pull completes. hf download fetches the real files directly, and
does not create a .git directory — neither is required for the model to
work; see §4.
4. Installation (wdnwebui)
Clone the repository into the model root so the folder holds both the weights and the worker manifest:
git clone https://huggingface.co/wardner/Qwen3-ASR /userdata/models/Qwen3-ASR
Then open the web UI → Compute → the card appears under "local compute" (the backend rescans the model root every ~2 s, no restart needed) → click "enable".
The directory name does not have to match the model name: the SDK follows
"the folder the manifest was found in", located through the
WDNAI_QWEN3_ASR_DIR environment variable declared in the manifest.
The card itself is Streaming Speech Recognition (qwen3-asr) in the app
grid. It accepts an uploaded audio file, a video file with an audio track,
or live audio from the board microphone; the recognized text is pushed to
the browser over WebSocket as each 1-second chunk completes.
5. SHA-256 verification
cd ./Qwen3-ASR
sha256sum -c SHA256SUMS
All entries must report OK before deployment.
6. Caveats and limits
See the notes field of qwen3-asr.json — it records the
per-model specifics (streaming strategy, capability layout, side effects).
The details that bite most often:
- This build is the streaming one. The model folder ships
encoder_online.*, notencoder.*, so the spec runs the streaming path (use_online_encoder=True, the SDK default). The encoder choice is a load-time decision — which RKNN file goes on the NPU — so it lives in the SDK spec, not in the manifest. Switching to the one-shot path needsencoder.1828.rknn+encoder.weightanduse_online_encoder=False. - Streaming strategy (aligned with the C++ demo
examples/Qwen3_ASR/cpp/online_main.cc): 1 s per chunk (ONLINE_CHUNK_MS=1000), at most 50 new tokens per chunk (ONLINE_MAX_NEW_TOKENS=50), the first 4 rounds commit nothing (ONLINE_ROLLBACK_ROUNDS=4), later rounds roll back the last 5 tokens (ONLINE_ROLLBACK_TOKENS=5). The last chunk commits everything. - Only the committed prefix is monotonic. The trailing "unconfirmed tail" is rewritten by the next round — that is by design, not a bug.
- The language head comes from the model. It emits
language <lang>first (token151704is the separator); the SDK returns the language separately to the front end. With little context the first round can report the wrong language (e.g.Chinesefor English speech) — the final language is taken from the committed token stream. - Audio length. The streaming demo caps at 60 s
(
ONLINE_MAX_AUDIO_SECONDS); the exported LLM context is 24k tokens (MAX_CONTEXT_LEN = 1024*24), which covers roughly 20 minutes of audio — the offline demo caps at 20 minutes (OFFLINE_MAX_AUDIO_SECONDS). infer_timeout_s=120. One inference is N serial LLM rounds (15 s of audio = 16 rounds) — the 60 s default budget is not enough.memory_estimate_mb=1600: encoder 375 MB + LLM 435 MB + embedding table 297 MB.- Input format. Any sample rate is resampled to 16 kHz mono. The mel
frame count must be truncated to
n_frame * hopexactly like the C demo does — without that truncation the model input width mismatches and the first round already diverges. - Do not "improve" the resampler. Linear interpolation is what matches the C demo byte-for-byte; swapping in a Kaiser-sinc filter drops words.
- Decoding must skip special tokens, otherwise
<|im_end|>leaks into the tail of the transcript. mel_128_filters.txtmust be in the same directory as the weights (128 mel × 202 fp32; the SDK truncates to the first 128 × 201 =N_FFT/2+1values).- The tokenizer comes from another repo:
llm.tokenizer.ggufis exported fromQwen/Qwen3-0.6B, not from the Qwen3-ASR repository. - Model not deployed → load raises
FileNotFoundError, which the worker's load flow turns into anerroredstate.
7. License
The upstream model is released under the Apache License 2.0. The files here are RKNN-converted derivatives, not unmodified upstream weights. See LICENSE and NOTICE.
- Downloads last month
- 358
We're not able to determine the quantization variants.
docker model run hf.co/wardner/Qwen3-ASR