Ministral 14B
Trinity Nano Preview
0.6B
- Context
- 33Ktokens
- Released
- 1y ago
- Size
- 8Bparameters
Qwen3 Reranker 8B is a reranking model from Alibaba, part of the newly launched Qwen3 Embedding series and built on the Qwen3 foundation model. Alibaba's announcement claims the series achieves leading results across text embedding and reranking benchmarks and that its rerankers sharpen search relevance in text retrieval. The model is built for retrieval work: given a query and candidate documents, it outputs relevance scores, supporting multilingual, cross-lingual, and code retrieval across over 100 languages.
How it compares
Size class
Small8BAn open-weight model of 5 to 25 billion parameters, the size that runs on a single consumer GPU.
Open models of a similar size from other labs
Vs. your stack
Add the models you use to see how this one compares.
Sign inAdditional source evidence
- Source model
- alibaba/qwen3-reranker-8b
- Source family
- qwen
- Source type
- reranking
- Last updated
- 2025-06-05
- Model context
- 33K
- Model output limit
- —
- Open weights
- Yes
- File attachments
- No
- Reasoning
- No
- Tool calling
- No
- Temperature control
- No
Downloadable weights
Complete source record (JSON)
{
"id": "alibaba/qwen3-reranker-8b",
"name": "Qwen3 Reranker 8B",
"description": "Multilingual query-document reranking model supporting more than 100 languages",
"family": "qwen",
"type": "reranking",
"last_updated": "2025-06-05",
"attachment": false,
"reasoning": false,
"tool_call": false,
"temperature": false,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 32768,
"output": 0
},
"release_date": "2025-06-05",
"open_weights": true,
"license": "Apache-2.0",
"weights": [
{
"label": "Hugging Face",
"url": "https://huggingface.co/Qwen/Qwen3-Reranker-8B"
}
]
}Frequently asked questions
Running locally
Qwen3 Reranker 8B needs an estimated 6.1 GB of VRAM at Q4, 10.3 GB at Q8, and 18.9 GB at FP16. These figures are estimates based on the 8B parameter count, covering weights plus about 15% and 0.5 GB for a short context.
Yes. A 24 GB GPU fits FP16 at an estimated 18.9 GB, a 16 GB or 12 GB GPU fits Q8 at an estimated 10.3 GB, and an 8 GB GPU fits Q4 at an estimated 6.1 GB.
Yes, at Q4, whose estimated VRAM is 6.1 GB. Assuming macOS lets the GPU use about 75% of unified memory, the smallest common Mac memory size that fits Q4 is an estimated 16 GB.
Privacy and data
Yes. Qwen3 Reranker 8B has open weights, so you can self-host it for full control of your data.
Capabilities and specs
The context window of Qwen3 Reranker 8B is 33K tokens (32,768).
No. Qwen3 Reranker 8B accepts text input only.
No, tool calling is not supported.
No, a reasoning or thinking mode is not supported.
Quality and benchmarks
The closest alternatives, matched on model kind and release date with no comparable price, are Sarvam-M by Sarvam AI, Gemma 3n E2B IT by Google DeepMind, and Gemma 3n E4B IT by Google DeepMind.
Sarvam-MQwen3 Reranker 8B vs Sarvam-MGemma 3n E2B ITQwen3 Reranker 8B vs Gemma 3n E2B ITGemma 3n E4B ITQwen3 Reranker 8B vs Gemma 3n E4B IT
Versions and lineage
Both models have a 33K token (32,768) context window. Qwen3 Reranker 8B was released on June 5, 2025, while Qwen3 Reranker 0.6B was released on August 26, 2026. Qwen3 Reranker 0.6B is priced at $0.00 input and $0.00 output per 1M tokens.
Licensing and access
Qwen3 Reranker 8B is open weights, licensed under Apache-2.0.
Yes. Qwen3 Reranker 8B uses the Apache 2.0 license, a permissive license that allows commercial use.
Basics
Qwen3 Reranker 8B was made by Qwen and released on June 5, 2025.
Qwen3 Reranker 8B has open weights.
Generated from Vertical Tabs data, updated November 11, 2026.
Discussion
No comments yet
Be the first to share your thoughts.