InferX Catalog | Qwen3-VL-32B-Instruct-FP8

Qwen3-VL-32B-Instruct-FP8

Qwen3-VL-32B-Instruct-FP8 represents the pinnacle of mid-sized multimodal intelligence from Alibaba Cloud (Qwen Team),

Qwen multimodal image2text

Log in to deploy

Metadata

Provider

Qwen

Modality

multimodal

API type

image2text

Source

huggingface / Qwen/Qwen3-VL-32B-Instruct-FP8

Created

2026-04-10 01:27:55 UTC

Updated

2026-04-10 01:27:55 UTC

Catalog version

1

Visibility

Published

Specifications

Parameters

32.00B

MoE

No

Max model length

10000

Image

vllm/vllm-openai:v0.16.0

Default Deploy Config

GPU count

1

vRAM

50000 MB

Summary

1xGPU 50000 MB

Recommended Use Cases

—

Model Spec

{
    "image": "vllm/vllm-openai:v0.16.0",
    "commands": [
        "--model",
        "Qwen/Qwen3-VL-32B-Instruct-FP8",
        "--disable-custom-all-reduce",
        "--trust-remote-code",
        "--max-model-len",
        "10000"
    ],
    "resources": {
        "GPU": {
            "Count": 1,
            "vRam": 50000
        }
    },
    "envs": [
        [
            "VLLM_USE_DEEP_GEMM",
            "0"
        ]
    ],
    "policy": {
        "Obj": {
            "min_replica": 0,
            "max_replica": 1,
            "standby_per_node": 1,
            "parallel": 50,
            "queue_len": 100,
            "queue_timeout": 30.0,
            "scalein_timeout": 1.0,
            "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": "What is in this image?",
        "apiType": "image2text",
        "dataUrl": "https://www.ilankelman.org/stopsigns/australia.jpg",
        "prompts": [],
        "loadingTimeout": 90
    }
}