InferX Catalog | Qwen3-ASR-1.7B

Qwen3-ASR-1.7B

The Qwen3-ASR family includes Qwen3-ASR-1.7B and Qwen3-ASR-0.6B, which support language identification and ASR for 52 languages and dialects. Both leverage large-scale speech training data and the strong audio understanding capability of their foundation model, Qwen3-Omni.

Qwen multimodal audio2text

Log in to deploy

Metadata

Provider

Qwen

Modality

multimodal

API type

audio2text

Source

huggingface / Qwen/Qwen3-ASR-1.7B

Created

2026-04-06 17:53:15 UTC

Updated

2026-04-06 17:53:15 UTC

Catalog version

1

Visibility

Published

Specifications

Parameters

—

MoE

No

Max model length

21000

Image

vllm/vllm-omni:v0.16.0

Default Deploy Config

GPU count

1

vRAM

12000 MB

Summary

1xGPU 12000 MB

Recommended Use Cases

—

Model Spec

{
    "image": "vllm/vllm-omni:v0.16.0",
    "commands": [
        "serve",
        "Qwen/Qwen3-ASR-1.7B",
        "--gpu-memory-utilization",
        "0.85",
        "--max-model-len",
        "21000"
    ],
    "resources": {
        "GPU": {
            "Count": 1,
            "vRam": 12000
        }
    },
    "envs": [],
    "policy": {
        "Obj": {
            "min_replica": 0,
            "max_replica": 1,
            "standby_per_node": 1,
            "parallel": 100,
            "queue_len": 100,
            "queue_timeout": 300.0,
            "scalein_timeout": 0.01,
            "scaleout_policy": {
                "WaitQueueRatio": {
                    "wait_ratio": 0.1
                }
            },
            "runtime_config": {
                "graph_sync": false
            }
        }
    },
    "sample_query": {
        "body": {
            "max_tokens": "200",
            "temperature": "0"
        },
        "path": "v1/chat/completions",
        "prompt": "Describe the sounds you hear in this audio.",
        "apiType": "audio2text",
        "dataUrl": "https://huggingface.co/kyutai/tts-voices/resolve/main/alba-mackenna/casual.wav",
        "prompts": [],
        "loadingTimeout": 90
    }
}