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 pipelines handlin…

nvidia/llama-nemotron-rerank-vl-1b-v2

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 pipelines handling charts, tables, infographics, and mixed-media documents. Functions as a cross-encoder that accepts text queries paired with image, text, or combined document inputs, delivering approximately 6-7% recall improvements over embedding-only baselines on visual document retrieval benchmarks.

Specifications & capabilities

Input
text, image
Output
rerank
Context window · tokens
10240
Capabilities
multimodal, rerank, visionInput
Description source ↗

Evaluations

1–2 / 2

MTEB

53.41%

source id
mteb
leaderboard name
MTEB Reranking
dimension
embedding
metric
Average task score
rank
267
sample size
192
aggregation scope
splits
  • test
subsets
  • english
  • french
  • german
  • italian
  • portuguese
  • spanish
task_count
32
task_names
  • Vidore3ComputerScienceBGEm3Reranking.v2
  • Vidore3ComputerScienceBM25sReranking.v2
  • Vidore3ComputerScienceNemotron35Embed8BReranking.v2
  • Vidore3ComputerScienceQwen3VLEmbedding2BReranking.v2
  • Vidore3EnergyBGEm3Reranking.v2
  • Vidore3EnergyBM25sReranking.v2
  • Vidore3EnergyNemotron35Embed8BReranking.v2
  • Vidore3EnergyQwen3VLEmbedding2BReranking.v2
  • Vidore3FinanceEnBGEm3Reranking.v2
  • Vidore3FinanceEnBM25sReranking.v2
  • Vidore3FinanceEnNemotron35Embed8BReranking.v2
  • Vidore3FinanceEnQwen3VLEmbedding2BReranking.v2
  • Vidore3FinanceFrBGEm3Reranking.v2
  • Vidore3FinanceFrBM25sReranking.v2
  • Vidore3FinanceFrNemotron35Embed8BReranking.v2
  • Vidore3FinanceFrQwen3VLEmbedding2BReranking.v2
  • Vidore3HrBGEm3Reranking.v2
  • Vidore3HrBM25sReranking.v2
  • Vidore3HrNemotron35Embed8BReranking.v2
  • Vidore3HrQwen3VLEmbedding2BReranking.v2
  • Vidore3IndustrialBGEm3Reranking.v2
  • Vidore3IndustrialBM25sReranking.v2
  • Vidore3IndustrialNemotron35Embed8BReranking.v2
  • Vidore3IndustrialQwen3VLEmbedding2BReranking.v2
  • Vidore3PharmaceuticalsBGEm3Reranking.v2
  • Vidore3PharmaceuticalsBM25sReranking.v2
  • Vidore3PharmaceuticalsNemotron35Embed8BReranking.v2
  • Vidore3PharmaceuticalsQwen3VLEmbedding2BReranking.v2
  • Vidore3PhysicsBGEm3Reranking.v2
  • Vidore3PhysicsBM25sReranking.v2
  • Vidore3PhysicsNemotron35Embed8BReranking.v2
  • Vidore3PhysicsQwen3VLEmbedding2BReranking.v2
sample_size
192
subset_count
6
subset_sample
  • english
  • french
  • german
  • italian
  • portuguese
  • spanish
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • b8a9987b05b75db5ad949c825cbcbf6eb7c48b5a
source model id
nvidia/llama-nemotron-rerank-vl-1b-v2
Original result ↗

MTEB

53.41%

source id
mteb
leaderboard name
MTEB Overall
dimension
embedding
metric
Average task score
rank
496
sample size
192
aggregation scope
splits
  • test
subsets
  • english
  • french
  • german
  • italian
  • portuguese
  • spanish
task_count
32
task_names
  • Vidore3ComputerScienceBGEm3Reranking.v2
  • Vidore3ComputerScienceBM25sReranking.v2
  • Vidore3ComputerScienceNemotron35Embed8BReranking.v2
  • Vidore3ComputerScienceQwen3VLEmbedding2BReranking.v2
  • Vidore3EnergyBGEm3Reranking.v2
  • Vidore3EnergyBM25sReranking.v2
  • Vidore3EnergyNemotron35Embed8BReranking.v2
  • Vidore3EnergyQwen3VLEmbedding2BReranking.v2
  • Vidore3FinanceEnBGEm3Reranking.v2
  • Vidore3FinanceEnBM25sReranking.v2
  • Vidore3FinanceEnNemotron35Embed8BReranking.v2
  • Vidore3FinanceEnQwen3VLEmbedding2BReranking.v2
  • Vidore3FinanceFrBGEm3Reranking.v2
  • Vidore3FinanceFrBM25sReranking.v2
  • Vidore3FinanceFrNemotron35Embed8BReranking.v2
  • Vidore3FinanceFrQwen3VLEmbedding2BReranking.v2
  • Vidore3HrBGEm3Reranking.v2
  • Vidore3HrBM25sReranking.v2
  • Vidore3HrNemotron35Embed8BReranking.v2
  • Vidore3HrQwen3VLEmbedding2BReranking.v2
  • Vidore3IndustrialBGEm3Reranking.v2
  • Vidore3IndustrialBM25sReranking.v2
  • Vidore3IndustrialNemotron35Embed8BReranking.v2
  • Vidore3IndustrialQwen3VLEmbedding2BReranking.v2
  • Vidore3PharmaceuticalsBGEm3Reranking.v2
  • Vidore3PharmaceuticalsBM25sReranking.v2
  • Vidore3PharmaceuticalsNemotron35Embed8BReranking.v2
  • Vidore3PharmaceuticalsQwen3VLEmbedding2BReranking.v2
  • Vidore3PhysicsBGEm3Reranking.v2
  • Vidore3PhysicsBM25sReranking.v2
  • Vidore3PhysicsNemotron35Embed8BReranking.v2
  • Vidore3PhysicsQwen3VLEmbedding2BReranking.v2
sample_size
192
subset_count
6
subset_sample
  • english
  • french
  • german
  • italian
  • portuguese
  • spanish
aggregation_method
mean-of-task-means
model_revision_count
1
model_revision_sample
  • b8a9987b05b75db5ad949c825cbcbf6eb7c48b5a
source model id
nvidia/llama-nemotron-rerank-vl-1b-v2
Original result ↗

Published prices

1–1 / 1

Provider specifications & access

1–1 / 1

llama-nemotron-rerank-vl-1b-v2 ↗

source model id
nvidia/llama-nemotron-rerank-vl-1b-v2
hosting
nvidia-catalog
platforms
  • NVIDIA Build / NIM
context length
10240
architecture
input_modalities
    output_modalities
      capabilities
      • rerank

      Safety assessments

      1–1 / 1

      Foundation Model Transparency Index 2025 ↗

      31 / 100

      source id
      stanfordFMTI
      canonical model id
      nvidia/llama-nemotron-rerank-vl-1b-v2
      taxonomy
      transparency
      hazard category
      transparency
      hazard name
      Organization transparency
      policy area
      Meta
      assessment type
      organization-level disclosure
      metric
      disclosure_score
      result label
      31 / 100
      sample size
      100
      updated at
      2026-11-09T18:08:05.227Z

      Source specification observations

      1–3 / 3

      nvidiaNim ↗

      field
      context_length
      value
      10240
      normalized value
      10240
      scope
      source-model
      source id
      nvidiaNim
      confidence
      1

      openRouter ↗

      field
      context_length
      value
      10240
      normalized value
      10240
      scope
      source-model
      source id
      openRouter
      confidence
      1

      openRouter ↗

      field
      max_output_tokens
      value
      9216
      normalized value
      9216
      scope
      source-model
      source id
      openRouter
      confidence
      1

      Original sources

      1–9 / 9

      DeepInfra · Model page ↗

      NVIDIA · Model page ↗

      OpenRouter · Model page ↗

      NVIDIA · Documentation ↗

      DeepInfra · Catalog ↗

      GitHub · Catalog ↗

      Hugging Face · Catalog ↗

      NVIDIA · Catalog ↗

      OpenRouter · Catalog ↗

      Source links

      1–5 / 5

      Explore more models

      Models

      Loading
      EVIDENCE / COMPARISON

      Benchmarks

      Explore every benchmark

      Loading

      These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.

      Install TensorTrace