Specialist modelOpen weights
LensVLM-9B
Apple
- Released
- -
- Data date
- October 3, 2026
LensVLM-9B is Apple’s vision-language model for long documents. It scans compressed visual representations of text and expands only the pages relevant to the current question. Visual processing therefore stays selective instead of loading every page at full resolution.
The 9-billion-parameter checkpoint runs locally with Transformers, vLLM, or SGLang under Apple’s Machine Learning Research Model License. The released demo supports 5x, 10x, and 15x compression. Its model card lists no hosted Inference Provider.
Specifications and access
| Specification | Value and source |
|---|---|
| Model ID | apple/LensVLM-9BSource |
| Model class | Vision-language model for selective context expansionSource |
| Parameters | 9 billionSource |
| Compression | 5x, 10x, or 15x input compressionSource |
| Access | Local execution, no hosted Inference Provider on Hugging FaceSource |
| License | Apple Machine Learning Research Model LicenseSource |
Measurements without matching peer values
These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.
LensVLM document QA accuracy, 5x (Show measurement, test conditions, and source)
- Source value
- 68.9
- Score
- 68.9
- Metric
- percent
- Unit
- %
- Category
- document-qa
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- LensVLM technical report
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Final LensVLM SFT+RL; 5x input compression; LLM-judge semantic accuracy; unweighted macro-average over seven datasets.
LensVLM document QA accuracy, 10x (Show measurement, test conditions, and source)
- Source value
- 62.1
- Score
- 62.1
- Metric
- percent
- Unit
- %
- Category
- document-qa
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- LensVLM technical report
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Final LensVLM SFT+RL; 10x input compression; LLM-judge semantic accuracy; unweighted macro-average over seven datasets.
LensVLM document QA accuracy, 15x (Show measurement, test conditions, and source)
- Source value
- 52.1
- Score
- 52.1
- Metric
- percent
- Unit
- %
- Category
- document-qa
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- LensVLM technical report
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Final LensVLM SFT+RL; 15x input compression; LLM-judge semantic accuracy; unweighted macro-average over seven datasets.
LensVLM evidence-page selection, 5x (Show measurement, test conditions, and source)
- Source value
- 76.8
- Score
- 76.8
- Metric
- percent
- Unit
- %
- Category
- document-retrieval
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- LensVLM technical report
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Final LensVLM SFT+RL; 5x input compression; ground-truth evidence-page hit.
LensVLM evidence-page selection, 10x (Show measurement, test conditions, and source)
- Source value
- 71
- Score
- 71
- Metric
- percent
- Unit
- %
- Category
- document-retrieval
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- LensVLM technical report
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Final LensVLM SFT+RL; 10x input compression; ground-truth evidence-page hit.
LensVLM evidence-page selection, 15x (Show measurement, test conditions, and source)
- Source value
- 52.1
- Score
- 52.1
- Metric
- percent
- Unit
- %
- Category
- document-retrieval
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- LensVLM technical report
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Final LensVLM SFT+RL; 15x input compression; ground-truth evidence-page hit.
LensVLM effective compression ratio, 5x (Show measurement, test conditions, and source)
- Source value
- 4.3
- Score
- 4.3
- Metric
- ratio
- Unit
- ratio
- Category
- compression
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- LensVLM technical report
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Final LensVLM SFT+RL; report ECR at 5x input compression.
LensVLM effective compression ratio, 10x (Show measurement, test conditions, and source)
- Source value
- 7.4
- Score
- 7.4
- Metric
- ratio
- Unit
- ratio
- Category
- compression
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- LensVLM technical report
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Final LensVLM SFT+RL; report ECR at 10x input compression.
LensVLM effective compression ratio, 15x (Show measurement, test conditions, and source)
- Source value
- 10.1
- Score
- 10.1
- Metric
- ratio
- Unit
- ratio
- Category
- compression
- Direction
- Higher is better
- Source type
- vendor-reported
- Evaluator
- LensVLM technical report
- Status
- active
- Retrieved at
- 2026-10-04
- Methodology
- Vendor-reported result. Final LensVLM SFT+RL; report ECR at 15x input compression.
Sources and data date
Every statement links to its underlying documentation or leaderboard.
| Type | Evidence and data date |
|---|---|
| Research status | Research date October 4, 2026. 2 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication. |
| Additional source | Apple LensVLM model card (retrieved October 3, 2026) · Apple LensVLM model card · Editorial description reviewed October 3, 2026 |
| Additional source | LensVLM research paper (retrieved October 3, 2026) · LensVLM research paper · Editorial description reviewed October 3, 2026 |
| Additional source | LensVLM technical report (retrieved October 4, 2026) |