Tīrtha · reranker
Nemotron 3.5 Rerank 8B
Scores how well each document answers a query and returns them best first, to sharpen search results before an answer is written. Send up to 100 documents in one call, up to 200,000 characters in total, query included. The query and each document together must fit the model's length limit.
Open weights · OpenMDW-1.1
by NVIDIA · OpenMDW-1.1 · open weights
About this model
- Languages
- 33 languages
- Model size
- 8.4B parameters
- Based on
- Ministral-3-8B-Instruct-2512-BF16
- Licence
- OpenMDW-1.1
How to use it
- Split long text into chunks; each model has its own length limit.
- Send the query and the candidate documents. Results come back best first, each with a score.
- Higher means more relevant; compare scores within one call.
How to call it
POST https://api.tirtha.ai/v1/rerank
{"model": "tirtha/nemotron-3.5-rerank-8b-bf16",
"query": "your question",
"documents": ["text one", "text two"]}
Start using Nemotron 3.5 Rerank 8B in 3 steps
Read the docs- Get an API keyFree to create. It works for every model in the catalogue.Get an API key
- Add creditsPay as you go. Credits work on every model.Add credits
- Point your client at TīrthaAny OpenAI-compatible client works. Use your key and this base URL.base_url: "https://api.tirtha.ai/v1" api_key: "YOUR_TIRTHA_KEY"