furiosa-ai/Qwen3-Reranker-8B

Model Overview

• Model Architecture: Qwen3
  • Input: Text (query-document pairs)
  • Output: Relevance Score
• Model Optimizations:
• Maximum Context Length: 8k tokens
  • Maximum Sequence Length: 8192 tokens
• Task Type: Reranking (relevance scoring)
• Intended Use Cases: Document reranking for retrieval-augmented generation (RAG) and search applications. Same as Qwen/Qwen3-Reranker-8B.
• Release Date: 01/20/2026
• Version: v2026.1
• License(s): Apache 2.0 License
• Supported Inference Engine(s): Furiosa LLM
• Supported Hardware Compatibility: FuriosaAI RNGD
• Preferred Operating System(s): Linux

Description:

This model is the pre-compiled version of the Qwen/Qwen3-Reranker-8B, which is a reranker model designed for scoring the relevance between queries and documents.

Usage

To run this model with Furiosa-LLM, follow the example command below after installing Furiosa-LLM and its prerequisites.

from furiosa_llm import LLM

llm = LLM.from_artifacts("furiosa-ai/Qwen3-Reranker-8B")
scores = llm.score([("query", "document1"), ("query", "document2")])