NVIDIA
Llama Nemotron Rerank VL 1B V2 (free)
nvidia/llama-nemotron-rerank-vl-1b-v2:free
Llama Nemotron Rerank VL 1B V2 is a 1.7B multimodal reranking model from NVIDIA. It evaluates the relevance of document images and text against user queries, designed for vision RAG...
RoleReorders passages by relevance to a query.
Free
In / M
Free
Out / M
10K
Context
- AA intelligence index
- Not on listing
- Inputs
- text, image
- Outputs
- rerank
- Tools
- Not listed
- Structured output
- Not listed
- Reasoning
- Not listed
- Weights
- Not verified
- Listed
- Jun 9, 2026
01 · What it's good at
Use it for this.
Three jobs this model is built for, and the ones to give to something else.
01
The last step of retrieval, after a vector search
Skip it for
- Generating the answer or the embedding
02 · Example tasks
What you'd actually send it.
Real inputs, not a benchmark. Paste one into the snippet below to try it.
You send
Rerank 50 vector hits to the top 5.
Why this model
Better passages, better answers.
03 · Your stack
Where it sits: Retrieve.
A reranker scores query-plus-passage pairs and returns an order. It is how RAG stops citing the wrong chunk.
Layer 01 · this model
Retrieve
Layer 02
Plan & reason
Layer 03
Do the volume
Layer 04
Gate & route
Layer 05
Draw, speak, listen
04 · Wire it up
One request. Paste and run.
Uses OpenRouter, so one key works for every model on this site. Set OPENROUTER_API_KEY and swap the prompt for one of the tasks above.
const res = await fetch("https://openrouter.ai/api/v1/rerank", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.OPENROUTER_API_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"model": "nvidia/llama-nemotron-rerank-vl-1b-v2:free",
"query": "Can I pay annually?",
"documents": [
"Billing runs monthly.",
"Annual plans are billed once a year."
],
"top_n": 2
}),
});
const { results } = await res.json();
for (const r of results) console.log(r.relevance_score.toFixed(3), r.document.text);
05 · Watch out for
Gotchas.
This lesson is generated from the model type and the live listing. It has not been hand-reviewed yet.
Check the listing for supported parameters before you rely on tools or structured output.
Sources
Lessons are written from each model's listing and the linked docs. We don't publish our own benchmark numbers; the index above is Artificial Analysis, as shown on OpenRouter.