All tasks / Models
Qwen3-Reranker-8B
Alibaba's Apache-2.0 open-weights text reranking models in 0.6B, 4B and 8B sizes.
- Alibaba (Qwen)
- Open weights
- Released Jun 2025
Where it ranks
- –Embeddings & retrievalModels · Local · no result on this task’s benchmarks yet
Facts
- Vendor
- Alibaba (Qwen)
- Family
- Qwen3-Reranker
- Released
- Jun 2025
- Weights
- Open: download and run it yourself
- License
- Apache-2.0
- Parameters
- 8.19B
- GPU memory to run locally
- ~17 GB (Derived: 8.19B parameters x 2 bytes = ~16.4 GB of bf16 weights.)
- Access
- Free (open weights)
- Last verified
- 19 Oct 2026
Benchmark results
| Benchmark | Result | Entry on the leaderboard | As of | Source |
|---|---|---|---|---|
| Agentset Reranker Leaderboard | 1,473 | Qwen3 Reranker 8BELO from GPT-5-judged pairwise comparisons; rank 8 of 12; win rate 51.2%; page says 'Last updated: February 15, 2026'. | 15 Feb 2026 | agentset.ai |
Where you can use it
- SvLLMVector databases & engines · Local