Qwen3 ASR 1.7B is Alibaba's speech to text model. The flagship model of the Qwen3-ASR family, offering multilingual speech recognition and language identification across 30 languages and 22 Chinese dialects with higher accuracy than its compact sibling.
Specifications
Canonical IDalibaba-qwen3-asr-1-7b
TypeSpeech to Text
StatusActive
CreatorAlibabaAlibaba
Input ModalitiesAudio
Output ModalitiesText
Parameters1.7B
HuggingFace Likes941
HuggingFace Downloads (30d)1,511,232
HuggingFace Downloads (all-time)8,611,447

Capabilities

Input1/5
Text·
Image·
Audio
Video·
PDF·
Output1/5
Text
Image·
Audio·
Video·
Embedding·
Capabilities0/13
Reasoning·
Adaptive Reasoning·
Function Calling·
Parallel Function Calling·
Structured Outputs·
Native JSON Schema·
Web Search·
URL Context·
Computer Use·
Code Execution·
File Search·
Prompt Caching·
Assistant Prefill·

Cheapest Instances to Run It

Cloud GPU instances that can host Qwen3 ASR 1.7B, ranked by cheapest on-demand price. The model needs about 4 GB of GPU memory at FP16 precision (estimated from its parameter count), so treat the fit as guidance rather than a guarantee.

All clouds
FP16 (full precision)
US Dollar ($)
Instance
Cloud
GPU
VRAM
Price
Cheapest region
Standard_NV4as_v4AzureAMD Radeon Instinct MI2516 GB$0.233/hr
g4ad.xlargeAWSRadeon Pro V5208 GB$0.379/hr
g5g.xlargeAWST4g16 GB$0.420/hr
7 more instances can run Qwen3 ASR 1.7B
Unlock the full ranked list and FP8 / INT4 quantization with a CloudPrice subscription.

Versions

VersionReleasedContextInput / 1MOutput / 1MStatus
Qwen3.8 2.4T A95B1.0M$2.00$6.00Available
Qwen Audio 3 Flash Realtime$0.413$4.13Available
Qwen Audio 3 Plus Realtime$0.688$5.50Available
Qwen Audio 3 Plus TTSAvailable
Qwen Image 3Available
Qwen Image 3.0 ProAvailable
EAGLE Qwen 2.5 3B InstructAvailable
DeepSeek R1 Qwen3 8BAvailable
Qwen3.8 27B1.0M$0.425$2.55Available
Qwen3.8 Max1.0M$1.65$4.99Available
Qwen3 ASR 1.7BCurrent

Model IDs

alibaba-qwen3-asr-1-7b
Qwen/Qwen3-ASR-1.7B