Rerankers

RankVicuna / RankLLaMA / RankZephyr (Castorini)

Rerankers that are language models: they receive the whole list of retrieved passages and reorder it as a list, instead of scoring passages one by one. Heavier than ordinary rerankers.

The last open version came out in Jan 2024. The family has not been updated for a long time: the model still works, but do not expect fixes or new sizes.

Developer
University of Waterloo, Castorini group, Canada
First release
Sep 2023
Latest release
Jan 2024
Sizes
7B – 13B
License
Commercial use with conditionsMIT (RankZephyr); RankVicuna and RankLLaMA — under their base model licenses
Russian
Not supported
Ready-made builds
GGUF
Running
On your own serverNeeds a GPU
Industries
Software development, Science and research, Documents and accounting
Deploy on your server Hugging Face GitHub

What it does

  • Reordering a long list of search results
  • Selecting sources for an AI assistant answer
  • Research comparisons of retrieval approaches

Where it is used

IT and software developmentResearch and scienceInternal knowledge bases

Hardware requirements

LaptopLaptop or regular PC, up to 8 GB of VRAM — smaller versions
fits
1 GPUOne GPU with 16–80 GB — mid-size versions
fits
ClusterServer with several GPUs — flagship versions
no versions

Versions

  1. RankZephyr 7B
  2. RankLLaMA 7B и 13B
  3. RankVicuna 7B

How to run it

On your own serverWeights are downloaded from Hugging Face and served with vLLM (load and API) or llama.cpp (modest hardware). The model then runs in a closed network with no per-request fees.Hugging Face
How much hardware you needCalculate the VRAM for the model size, context length and number of concurrent requests.Open the hardware calculator

I can set this up end to end: pick the model size, deploy it on your server and connect it to your systems. Quantization compresses a model so it takes less video memory and runs on more modest hardware. Answers change slightly, so quality is checked on your own examples.

Frequently asked questions

Can RankVicuna / RankLLaMA / RankZephyr (Castorini) be used in a commercial project?

With conditions. License: MIT (RankZephyr); RankVicuna and RankLLaMA — under their base model licenses. Restrictions vary — region, company revenue, attribution requirements. Have a lawyer check the terms before a commercial launch.

What hardware does RankVicuna / RankLLaMA / RankZephyr (Castorini) need?

At minimum: One GPU with 16–80 GB — mid-size versions. Without a GPU the model is not practical. You can calculate the exact VRAM for your model size and context in the hardware calculator.

Does RankVicuna / RankLLaMA / RankZephyr (Castorini) support Russian?

No. The developer documentation lists the supported languages and Russian is not among them.

Where can I download RankVicuna / RankLLaMA / RankZephyr (Castorini) and what does it cost?

The RankVicuna / RankLLaMA / RankZephyr (Castorini) weights are open and free to download. You only pay for the hardware it runs on and for the setup. Source links are at the bottom of this page.

How I deploy it for clients

  1. SelectionI pick the model size for your task and hardware and test it on your examples.
  2. DeploymentI deploy it on your server or in a closed network and provide an API.
  3. Fine-tuningI fine-tune it on your data (LoRA) or connect a knowledge base — whichever is cheaper for the task.
  4. IntegrationI connect it to your CRM, ERP, bot, website or team chat and set up monitoring.

Similar models

RerankersBGE RerankerBAAI (Beijing Academy of Artificial Intelligence) · ChinaCommercial use allowed

Rerankers: they take passages found by search and reorder them by how well they actually match the question. v2-m3 is multilingual and lightweight, often paired with bge-m3.

DetailsRerankersCross-Encoder MS MARCO (MiniLM, TinyBERT, mMARCO)UKP Lab and the Sentence Transformers community · GermanyCommercial use allowed

The most downloaded open rerankers: a tiny model reads a question-passage pair and scores how well they match. The multilingual mMARCO version covers Russian.

DetailsRerankerszerank (ZeroEntropy)ZeroEntropy · USACommercial use allowed

Rerankers built on Qwen3. The model card lists the target domains — finance, law, code, medicine, science; the stated language is English.

Details

Source: huggingface.co/castorini/rank_zephyr_7b_v1_full. Data checked against the model card on 22 Sep 2026. Have a lawyer review the license before commercial launch.