InferX Catalog | Inferx-bundle-Qwen3.6-35B-A3B-FP8-Qwen3-Embedding-0.6B-Qwen3-Reranker-0.6B

Inferx-bundle-Qwen3.6-35B-A3B-FP8-Qwen3-Embedding-0.6B-Qwen3-Reranker-0.6B

this is a bundle of Qwen3.6-35B-A3B-FP8, Qwen3-Embedding-0.6B and Qwen3-Reranker-0.6B

Qwen text text2text

Log in to deploy

This is a bundle of Qwen3.6-35B-A3B-FP8, Qwen3-Embedding-0.6B and Qwen3-Reranker-0.6B designed for RAG sceanrio. Then endpont is: 1. LLM: https://model.inferx.net/funccall/<Tenant>/<Namespace>/<ModalName>/v1 2. Rerank: https://model.inferx.net/funccall/<Tenant>/<Namespace>/<ModalName>/m2/v1 3. Embedding: https://model.inferx.net/funccall/<Tenant>/<Namespace>/<ModalName>/m3/v1

Metadata

Provider

Qwen

Modality

text

API type

text2text

Source

huggingface / Qwen/Qwen3.6-35B-A3B-FP8

Created

2026-04-21 16:34:22 UTC

Updated

2026-04-21 16:34:22 UTC

Catalog version

1

Visibility

Published

Specifications

Parameters

36.00B

MoE

No

Max model length

—

Image

inferx/inferx-bundle:v0.19.0

Default Deploy Config

GPU count

1

vRAM

66000 MB

Summary

1xGPU 66000 MB

Recommended Use Cases

—

Model Spec

{
    "image": "inferx/inferx-bundle:v0.19.0",
    "commands": [
        "python3 -m vllm.entrypoints.openai.api_server --model Qwen/Qwen3-Reranker-0.6B --gpu-memory-utilization 0.05 --enable-sleep-mode --max-model-len 1200 --runner pooling --port 8002",
        "python3 -m vllm.entrypoints.openai.api_server --model Qwen/Qwen3-Embedding-0.6B --gpu-memory-utilization 0.05 --enable-sleep-mode --max-model-len 1200 --runner pooling --port 8003",
        "python3 -m vllm.entrypoints.openai.api_server --model Qwen/Qwen3.6-35B-A3B-FP8 --gpu-memory-utilization 0.80 --enable-sleep-mode --max-model-len 32000 --max-num-batched-tokens 4096  --enable-chunked-prefill --enable-prefix-caching --port 8001"
    ],
    "resources": {
        "GPU": {
            "Count": 1,
            "vRam": 66000
        }
    },
    "envs": [
        [
            "VLLM_SERVER_DEV_MODE",
            "1"
        ],
        [
            "VLLM_USE_DEEP_GEMM",
            "0"
        ]
    ],
    "sample_query": {
        "body": {
            "stream": "true",
            "messages": [
                {
                    "role": "system",
                    "content": "You are a helpful assistant."
                },
                {
                    "role": "user",
                    "content": "Seattle is a"
                }
            ],
            "max_tokens": "100",
            "temperature": "0"
        },
        "path": "v1/chat/completions",
        "prompt": "Seattle is a",
        "apiType": "text2text",
        "dataUrl": "",
        "prompts": [],
        "raw_query": true,
        "max_tokens": 100,
        "loadingTimeout": 90
    }
}