Qwen3-Reranker-0.6B-W8A16

Qwen3-Reranker-0.6B

<p align="center"> <img src="https://qianwen-res.oss-accelerate-overseas.aliyuncs.com/logo_qwen3.png" width="400"/> <p>

Highlights

The Qwen3 Embedding model series is the latest proprietary model of the Qwen family, specifically designed for text embedding and ranking tasks. Building upon the dense foundational models of the Qwen3 series, it provides a comprehensive range of text embeddings and reranking models in various sizes (0.6B, 4B, and 8B). This series inherits the exceptional multilingual capabilities, long-text understanding, and reasoning skills of its foundational model. The Qwen3 Embedding series represents significant advancements in multiple text embedding and ranking tasks, including text retrieval, code retrieval, text classification, text clustering, and bitext mining.

Exceptional Versatility: The embedding model has achieved state-of-the-art performance across a wide range of downstream application evaluations. The 8B size embedding model ranks No.1 in the MTEB multilingual leaderboard (as of June 5, 2025, score 70.58), while the reranking model excels in various text retrieval scenarios.

Comprehensive Flexibility: The Qwen3 Embedding series offers a full spectrum of sizes (from 0.6B to 8B) for both embedding and reranking models, catering to diverse use cases that prioritize efficiency and effectiveness. Developers can seamlessly combine these two modules. Additionally, the embedding model allows for flexible vector definitions across all dimensions, and both embedding and reranking models support user-defined instructions to enhance performance for specific tasks, languages, or scenarios.

Multilingual Capability: The Qwen3 Embedding series offer support for over 100 languages, thanks to the multilingual capabilites of Qwen3 models. This includes various programming languages, and provides robust multilingual, cross-lingual, and code retrieval capabilities.

Model Overview

Qwen3-Reranker-0.6B has the following features:

• Model Type: Text Reranking
• Supported Languages: 100+ Languages
• Number of Paramaters: 0.6B
• Context Length: 32k

For more details, including benchmark evaluation, hardware requirements, and inference performance, please refer to our blog, GitHub.

Qwen3 Embedding Series Model list

Note:

• MRL Support indicates whether the embedding model supports custom dimensions for the final embedding.
• Instruction Aware notes whether the embedding or reranking model supports customizing the input instruction according to different tasks.
• Our evaluation indicates that, for most downstream tasks, using instructions (instruct) typically yields an improvement of 1% to 5% compared to not using them. Therefore, we recommend that developers create tailored instructions specific to their tasks and scenarios. In multilingual contexts, we also advise users to write their instructions in English, as most instructions utilized during the model training process were originally written in English.

Usage

Using Sentence Transformers

Install Sentence Transformers:

pip install sentence_transformers

from sentence_transformers import CrossEncoder

model = CrossEncoder("Qwen/Qwen3-Reranker-0.6B")

query = "What is the capital of China?"
documents = [
    "The capital of China is Beijing.",
    "Gravity is a force that attracts two bodies towards each other. It gives weight to physical objects and is responsible for the movement of planets around the sun.",
]

pairs = [(query, doc) for doc in documents]
scores = model.predict(pairs)
print(scores)
# [ 7.625 -11.375]

rankings = model.rank(query, documents)
print(rankings)
# [{'corpus_id': 0, 'score': 7.625}, {'corpus_id': 1, 'score': -11.375}]

By default, scores are raw logit differences. To get 0-1 probability scores, pass a Sigmoid activation function:

scores = model.predict([(query, doc) for doc in documents], activation_fn=torch.nn.Sigmoid())

The model uses a default prompt "query" which injects the instruction "Given a web search query, retrieve relevant passages that answer the query" into the chat template. You can provide a custom instruction via the prompts parameter:

model = CrossEncoder(
    "Qwen/Qwen3-Reranker-0.6B",
    prompts={"classification": "Classify whether the document matches the query topic"},
    default_prompt_name="classification",
)

Using Transformers

With Transformers versions earlier than 4.51.0, you may encounter the following error:

KeyError: 'qwen3'

# Requires transformers>=4.51.0
import torch
from transformers import AutoModel, AutoTokenizer, AutoModelForCausalLM

def format_instruction(instruction, query, doc):
    if instruction is None:
        instruction = 'Given a web search query, retrieve relevant passages that answer the query'
    output = "<Instruct>: {instruction}\n<Query>: {query}\n<Document>: {doc}".format(instruction=instruction,query=query, doc=doc)
    return output

def process_inputs(pairs):
    inputs = tokenizer(
        pairs, padding=False, truncation='longest_first',
        return_attention_mask=False, max_length=max_length - len(prefix_tokens) - len(suffix_tokens)
    )
    for i, ele in enumerate(inputs['input_ids']):
        inputs['input_ids'][i] = prefix_tokens + ele + suffix_tokens
    inputs = tokenizer.pad(inputs, padding=True, return_tensors="pt", max_length=max_length)
    for key in inputs:
        inputs[key] = inputs[key].to(model.device)
    return inputs

@torch.no_grad()
def compute_logits(inputs, **kwargs):
    batch_scores = model(**inputs).logits[:, -1, :]
    true_vector = batch_scores[:, token_true_id]
    false_vector = batch_scores[:, token_false_id]
    batch_scores = torch.stack([false_vector, true_vector], dim=1)
    batch_scores = torch.nn.functional.log_softmax(batch_scores, dim=1)
    scores = batch_scores[:, 1].exp().tolist()
    return scores

tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-Reranker-0.6B", padding_side='left')
model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-Reranker-0.6B").eval()
# We recommend enabling flash_attention_2 for better acceleration and memory saving.
# model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-Reranker-0.6B", torch_dtype=torch.float16, attn_implementation="flash_attention_2").cuda().eval()
token_false_id = tokenizer.convert_tokens_to_ids("no")
token_true_id = tokenizer.convert_tokens_to_ids("yes")
max_length = 8192

prefix = "<|im_start|>system\nJudge whether the Document meets the requirements based on the Query and the Instruct provided. Note that the answer can only be \"yes\" or \"no\".<|im_end|>\n<|im_start|>user\n"
suffix = "<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n"
prefix_tokens = tokenizer.encode(prefix, add_special_tokens=False)
suffix_tokens = tokenizer.encode(suffix, add_special_tokens=False)
        
task = 'Given a web search query, retrieve relevant passages that answer the query'

queries = ["What is the capital of China?",
    "Explain gravity",
]

documents = [
    "The capital of China is Beijing.",
    "Gravity is a force that attracts two bodies towards each other. It gives weight to physical objects and is responsible for the movement of planets around the sun.",
]

pairs = [format_instruction(task, query, doc) for query, doc in zip(queries, documents)]

# Tokenize the input texts
inputs = process_inputs(pairs)
scores = compute_logits(inputs)

print("scores: ", scores)

vLLM Usage

# Requires vllm>=0.8.5
import logging
from typing import Dict, Optional, List

import json
import logging

import torch

from transformers import AutoTokenizer, is_torch_npu_available
from vllm import LLM, SamplingParams
from vllm.distributed.parallel_state import destroy_model_parallel
import gc
import math
from vllm.inputs.data import TokensPrompt


        
def format_instruction(instruction, query, doc):
    text = [
        {"role": "system", "content": "Judge whether the Document meets the requirements based on the Query and the Instruct provided. Note that the answer can only be \"yes\" or \"no\"."},
        {"role": "user", "content": f"<Instruct>: {instruction}\n\n<Query>: {query}\n\n<Document>: {doc}"}
    ]
    return text

def process_inputs(pairs, instruction, max_length, suffix_tokens):
    messages = [format_instruction(instruction, query, doc) for query, doc in pairs]
    messages =  tokenizer.apply_chat_template(
        messages, tokenize=True, add_generation_prompt=False, enable_thinking=False
    )
    messages = [ele[:max_length] + suffix_tokens for ele in messages]
    messages = [TokensPrompt(prompt_token_ids=ele) for ele in messages]
    return messages

def compute_logits(model, messages, sampling_params, true_token, false_token):
    outputs = model.generate(messages, sampling_params, use_tqdm=False)
    scores = []
    for i in range(len(outputs)):
        final_logits = outputs[i].outputs[0].logprobs[-1]
        token_count = len(outputs[i].outputs[0].token_ids)
        if true_token not in final_logits:
            true_logit = -10
        else:
            true_logit = final_logits[true_token].logprob
        if false_token not in final_logits:
            false_logit = -10
        else:
            false_logit = final_logits[false_token].logprob
        true_score = math.exp(true_logit)
        false_score = math.exp(false_logit)
        score = true_score / (true_score + false_score)
        scores.append(score)
    return scores

number_of_gpu = torch.cuda.device_count()
tokenizer = AutoTokenizer.from_pretrained('Qwen/Qwen3-Reranker-0.6B')
model = LLM(model='Qwen/Qwen3-Reranker-0.6B', tensor_parallel_size=number_of_gpu, max_model_len=10000, enable_prefix_caching=True, gpu_memory_utilization=0.8)
tokenizer.padding_side = "left"
tokenizer.pad_token = tokenizer.eos_token
suffix = "<|im_end|>\n<|im_start|>assistant\n<think>\n\n</think>\n\n"
max_length=8192
suffix_tokens = tokenizer.encode(suffix, add_special_tokens=False)
true_token = tokenizer("yes", add_special_tokens=False).input_ids[0]
false_token = tokenizer("no", add_special_tokens=False).input_ids[0]
sampling_params = SamplingParams(temperature=0, 
    max_tokens=1,
    logprobs=20, 
    allowed_token_ids=[true_token, false_token],
)

        
task = 'Given a web search query, retrieve relevant passages that answer the query'
queries = ["What is the capital of China?",
    "Explain gravity",
]
documents = [
    "The capital of China is Beijing.",
    "Gravity is a force that attracts two bodies towards each other. It gives weight to physical objects and is responsible for the movement of planets around the sun.",
]

pairs = list(zip(queries, documents))
inputs = process_inputs(pairs, task, max_length-len(suffix_tokens), suffix_tokens)
scores = compute_logits(model, inputs, sampling_params, true_token, false_token)
print('scores', scores)

destroy_model_parallel()

📌 Tip: We recommend that developers customize the instruct according to their specific scenarios, tasks, and languages. Our tests have shown that in most retrieval scenarios, not using an instruct on the query side can lead to a drop in retrieval performance by approximately 1% to 5%.

Evaluation

Note:

• Evaluation results for reranking models. We use the retrieval subsets of MTEB(eng, v2), MTEB(cmn, v1), MMTEB and MTEB (Code), which are MTEB-R, CMTEB-R, MMTEB-R and MTEB-Code.
• All scores are our runs based on the top-100 candidates retrieved by dense embedding model Qwen3-Embedding-0.6B.

Citation

If you find our work helpful, feel free to give us a cite.

@article{qwen3embedding,
  title={Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models},
  author={Zhang, Yanzhao and Li, Mingxin and Long, Dingkun and Zhang, Xin and Lin, Huan and Yang, Baosong and Xie, Pengjun and Yang, An and Liu, Dayiheng and Lin, Junyang and Huang, Fei and Zhou, Jingren},
  journal={arXiv preprint arXiv:2506.05176},
  year={2025}
}

Quantization: W8A16 + calibrated FP8 KV

This is a W8A16 quantization of Qwen/Qwen3-Reranker-0.6B produced with llm-compressor:

• Weights: INT8, group-128 symmetric, via GPTQModifier (scheme="W8A16")
• KV cache: FP8, per-tensor symmetric, calibrated on 512 domain-matched samples
• Activations: unquantized (FP16/BF16), weight-only
• Output format: compressed-tensors, auto-detected by vLLM from config.json

The full recipe is in recipe.yaml. Calibration set drawn from the deployment domain (personality rules corpus, auto-memory entries, and real agent session query+document pairs) — 512 samples, unsupervised. Served and validated on a Jetson Orin Nano Super (Ampere/Tegra) with vLLM 0.27.1 and on RTX 3090 Ti with vLLM 0.28.0.

The upstream model card above describes the base model.