# Nanonets-OCR2-3B by Nanonets: Open-Weight Model
Source: https://savrn.com/models/nanonets-ocr2-3b
Markdown alternate of the page above; the site index is https://savrn.com/llms.txt

---

## Runs On

What it takes to serve Nanonets-OCR2-3B (3.8B parameters): the memory its weights need at each precision, and the cheapest way to rent enough data-center GPUs to hold them.

| Precision | Weights | Memory needed | Cheapest setup | Per hour | Also fits |
| --- | --- | --- | --- | --- | --- |
| 16-bit | 7.5 GB | 9.0 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 |
| 8-bit | 3.8 GB | 4.5 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 |
| 4-bit | 1.9 GB | 2.3 GB | 1x [MI300X](https://savrn.com/ai-index/pricing/gpus/mi300x) (192 GB) Vultr | $1.85 | [1x H100](https://savrn.com/ai-index/pricing/gpus/h100) $1.99 · [1x MI325X](https://savrn.com/ai-index/pricing/gpus/mi325x) $2.00 |

Memory is the weights at that precision plus 20% for the runtime and a short context; a long context needs more. Prices are the lowest on-demand hourly rates in the [SAVRN Index](https://savrn.com/ai-index/pricing/gpus), read Oct 23, 2026.

[Nanonets-OCR2-3B on every accelerator the SAVRN Index prices, at every precision](https://savrn.com/models/nanonets-ocr2-3b/gpus)

## SAVRN's Notes on Nanonets-OCR2-3B

Nanonets-OCR2 converts document images into structured markdown and tags content by meaning, so tables, equations, signatures and checkboxes come out labeled, not as loose text. It is built on Qwen2.5-VL at 3.8B parameters with a 128,000-token context.

The repository states no license, so treat it as research use until Nanonets publishes terms. It needs about 9 GB at 16-bit. For document pipelines, its structured output can save a lot of downstream parsing.

SAVRN Research, 2026-09-18

## Model Card

Nanonets-OCR2: A model for transforming documents into structured markdown with intelligent content recognition and semantic tagging Nanonets-OCR2 by Nanonets is a family of powerful, state-of-the-art image-to-markdown OCR models that go far beyond traditional text extraction. It transforms documents into structured markdown with intelligent content recognition and semantic tagging, making it ideal for downstream processing by Large Language Models (LLMs). Nanonets-OCR2 is packed with features designed to handle complex documents with ease: 1. Start the vLLM server. Check out Docstrange for more details. 1. Increasing the image resolution will improve model's performance. 2. For complex…

Excerpt from the card by Nanonets.

## Configuration

Architecture

Qwen2_5_VLForConditionalGeneration

Context length (tokens)

128,000

Layers

36

Hidden size

2,048

Feed-forward size

11,008

Attention heads

16

Key/value heads

2

Vocabulary size

151,936

Sliding window (tokens)

32,768

RoPE base

1e+06

Stored precision

bfloat16

Model type

qwen2_5_vl

## Identity and Version

Repository

nanonets/Nanonets-OCR2-3B

Publisher

Nanonets

Task

Image and text to text

Modality

Image and text

Library

transformers

Parameters

3.8B parameters

Languages

Not stated by the source

Revision

c3886ff00bb037ce7da24988c9eafaf1fe2bed72

First published

2025-10-13

Last updated

2025-10-16

## Files and Weights

17 files, 7.5 GB in total. The weights are 2 files totalling 7.5 GB in safetensors.

Weights2 files · 7.5 GB

Configuration7 files · 72.0 KB

Tokenizer4 files · 15.9 MB

Documentation1 file · 16.1 KB

Other2 files · 1.4 KB

Repository1 file · 1.6 KB

Every file

| File | Type | Size | SHA-256 |
| --- | --- | --- | --- |
| model-00001-of-00002.safetensors | Weights | 5.0 GB | 84c3306e843b |
| model-00002-of-00002.safetensors | Weights | 2.5 GB | 8002ad9c8c62 |
| added_tokens.json | Configuration | 605 B | — |
| config.json | Configuration | 3.4 KB | — |
| generation_config.json | Configuration | 214 B | — |
| model.safetensors.index.json | Configuration | 65.5 KB | — |
| preprocessor_config.json | Configuration | 791 B | — |
| special_tokens_map.json | Configuration | 613 B | — |
| video_preprocessor_config.json | Configuration | 907 B | — |
| README.md | Documentation | 16.1 KB | — |
| Modelfile | Other | 424 B | — |
| chat_template.jinja | Other | 1.0 KB | — |
| .gitattributes | Repository | 1.6 KB | — |
| merges.txt | Tokenizer | 1.7 MB | — |
| tokenizer.json | Tokenizer | 11.4 MB | 9c5ae00e602b |
| tokenizer_config.json | Tokenizer | 4.8 KB | — |
| vocab.json | Tokenizer | 2.8 MB | — |

## License and Download

License

Not stated by the source

Access

Open weights, no gate

Download size

7.5 GB

[Download from Nanonets](https://huggingface.co/nanonets/Nanonets-OCR2-3B)

Released by Nanonets through its official repository on Hugging Face.

## Built From

- Derived from [Qwen/Qwen2.5-VL-3B-Instruct](https://savrn.com/models/qwen2-5-vl-3b-instruct)

## Evaluations

Each result is shown as reported, with the conditions its reporter stated. None is a SAVRN measurement. A comparison lines two results up only when their configuration, unit and setup are all stated and identical.

| Benchmark | Conditions | Result | Reported by | Revision | Date |
| --- | --- | --- | --- | --- | --- |
| Delores-Lin/MDPBench | Task arMetric arComparison conditions not established | 60.2 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task deMetric deComparison conditions not established | 76.7 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task digitalMetric digitalComparison conditions not established | 79.2 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task enMetric enComparison conditions not established | 76.4 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task esMetric esComparison conditions not established | 61.8 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task frMetric frComparison conditions not established | 66.1 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task hiMetric hiComparison conditions not established | 59.2 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task idMetric idComparison conditions not established | 68.4 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task itMetric itComparison conditions not established | 78.5 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task jpMetric jpComparison conditions not established | 52.1 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task koMetric koComparison conditions not established | 54.7 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task latinMetric latinComparison conditions not established | 71.4 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task nlMetric nlComparison conditions not established | 74.1 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task non_latinMetric non_latinComparison conditions not established | 56.2 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task overallMetric overallComparison conditions not established | 64.2 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task photographedMetric photographedComparison conditions not established | 59.3 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task ptMetric ptComparison conditions not established | 74.2 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task ruMetric ruComparison conditions not established | 45.5 | Not named Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| Delores-Lin/MDPBench | Task viMetric viComparison conditions not established | 66 | [MDPBench leaderboard](https://huggingface.co/datasets/Delores-Lin/MDPBench) Reported by a third party | Evaluated revision not stated | 2026-07-03 |
| allenai/olmOCR-bench | Task arxiv_mathMetric arxiv_mathComparison conditions not established | 75.4 | [OlmOCR-Bench Github repository](https://github.com/allenai/olmocr/tree/main/olmocr/bench) Reported by a third party | Evaluated revision not stated | 2026-02-19 |
| allenai/olmOCR-bench | Task headers_footersMetric headers_footersComparison conditions not established | 32.1 | [OlmOCR-Bench Github repository](https://github.com/allenai/olmocr/tree/main/olmocr/bench) Reported by a third party | Evaluated revision not stated | 2026-02-19 |
| allenai/olmOCR-bench | Task long_tiny_textMetric long_tiny_textComparison conditions not established | 93 | [OlmOCR-Bench Github repository](https://github.com/allenai/olmocr/tree/main/olmocr/bench) Reported by a third party | Evaluated revision not stated | 2026-02-19 |
| allenai/olmOCR-bench | Task multi_columnMetric multi_columnComparison conditions not established | 81.9 | [OlmOCR-Bench Github repository](https://github.com/allenai/olmocr/tree/main/olmocr/bench) Reported by a third party | Evaluated revision not stated | 2026-02-19 |
| allenai/olmOCR-bench | Task old_scansMetric old_scansComparison conditions not established | 40.9 | [OlmOCR-Bench Github repository](https://github.com/allenai/olmocr/tree/main/olmocr/bench) Reported by a third party | Evaluated revision not stated | 2026-02-19 |
| allenai/olmOCR-bench | Task old_scans_mathMetric old_scans_mathComparison conditions not established | 46.1 | [OlmOCR-Bench Github repository](https://github.com/allenai/olmocr/tree/main/olmocr/bench) Reported by a third party | Evaluated revision not stated | 2026-02-19 |
| allenai/olmOCR-bench | Task overallMetric overallComparison conditions not established | 69.5 | [OlmOCR-Bench Github repository](https://github.com/allenai/olmocr/tree/main/olmocr/bench) Reported by a third party | Evaluated revision not stated | 2026-02-19 |
| allenai/olmOCR-bench | Task table_testsMetric table_testsComparison conditions not established | 86.8 | [OlmOCR-Bench Github repository](https://github.com/allenai/olmocr/tree/main/olmocr/bench) Reported by a third party | Evaluated revision not stated | 2026-02-19 |

## Memory Requirements

| Precision | Weights in memory |
| --- | --- |
| As published | 7.5 GB |
| 16-bit | 7.5 GB |
| 8-bit | 3.8 GB |
| 4-bit | 1.9 GB |

Weights only, from the published parameter count; the key-value cache and runtime add to this.

## Compare Nanonets-OCR2-3B

- [Nanonets-OCR2-3B vs Qwen2.5-VL-3B-Instruct](https://savrn.com/models/compare/nanonets-ocr2-3b-vs-qwen2-5-vl-3b-instruct)

## Questions About Nanonets-OCR2-3B

### How much GPU memory does Nanonets-OCR2-3B need?

About 9 GB at 16-bit and 2.3 GB at 4-bit: the weights (3.8B parameters) plus a working margin. A long context needs more.

### What is the cheapest GPU to run Nanonets-OCR2-3B on?

At 16-bit, 1x MI300X from $1.85 an hour; at 4-bit, 1x MI300X from $1.85 an hour, at the lowest on-demand prices the SAVRN Index lists.

### What is Nanonets-OCR2-3B's context length?

128,000 tokens, from the maximum position embeddings in its published configuration.

## Similar Models

Model · Image and text to text

### [Qwen2.5-VL-3B-Instruct](https://savrn.com/models/qwen2-5-vl-3b-instruct)

[Qwen](https://savrn.com/model-publishers/qwen)

licensename: qwen-research licenselink: https://huggingface.co/Qwen/Qwen2.5-VL-3B-Instruct/blob/main/LICENSE pipelinetag: image-text-to-text - multimodal libraryname: transformers In the past five months since Qwen2-VL’s release, numerous developers have built new models on the Qwen2-VL vision-language models, providing us with valuable feedback. During this period, we focused on building more useful vision-language models. Today, we are excited to introduce the latest addition to the Qwen family: Qwen2.5-VL. Understanding long videos and capturing events: Qwen2.5-VL can comprehend videos of over 1 hour, and this time it has a new ability of cpaturing event by pinpointing the relevant video…

Open weights 3.8B parameters 128,000 tokens transformers

[View model](https://savrn.com/models/qwen2-5-vl-3b-instruct)

Model · Image and text to text

### [swiftvln-satnav-3b-1ep-f32s4-overlap0-pf-h8-pool-s2-noembed](https://savrn.com/models/swiftvln-satnav-3b-1ep-f32s4-overlap0-pf-h8-pool-s2-noembed)

[Jiajun Jiang](https://savrn.com/model-publishers/eku127)

SwiftVLN: training and evaluation code for these checkpoints. - SatNav: satellite-image navigation environments, datasets, and evaluation tools. This checkpoint is designed for SwiftVLN on SatNav, the continuous-state vision-and-language navigation benchmark over satellite imagery. It predicts short navigation-action sequences from an instruction, the current RGB window, and sampled visual memory. - swiftvln-satnav: SwiftVLN trained and evaluated on SatNav - 3b: Qwen2.5-VL 3B backbone - 1ep: trained for one epoch - f32s4: uses a 32-frame window and predicts four actions - overlap0: uses non-overlapping training windows - pf-h8: uses per-frame memory with up to eight history frames…

Open weights 3.8B parameters 128,000 tokens transformers

[View model](https://savrn.com/models/swiftvln-satnav-3b-1ep-f32s4-overlap0-pf-h8-pool-s2-noembed)

Model · Image and text to text

### [blip2-opt-2.7b](https://savrn.com/models/blip2-opt-2-7b)

[Salesforce AI Research](https://savrn.com/model-publishers/salesforce)

BLIP-2 model, leveraging OPT-2.7b (a large language model with 2.7 billion parameters). It was introduced in the paper BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models by Li et al. and first released in this repository. Disclaimer: The team releasing BLIP-2 did not write a model card for this model so this model card has been written by the Hugging Face team. BLIP-2 consists of 3 models: a CLIP-like image encoder, a Querying Transformer (Q-Former) and a large language model. The authors initialize the weights of the image encoder and large language model from pre-trained checkpoints and keep them frozen while training the Querying…

Open weights mit 3.7B parameters transformers

[View model](https://savrn.com/models/blip2-opt-2-7b)

Model · Image and text to text

### [LocateAnything-3B](https://savrn.com/models/locateanything-3b)

[NVIDIA](https://savrn.com/model-publishers/nvidia)

LocateAnything is a vision-language model for fast and high-quality visual grounding, enabling precise object localization, dense detection, and point-based localization across diverse domains in both Enterprise Intelligence and Physical AI. The model adopts a generalist design, supporting tasks such as referring expression grounding, multi-object detection, GUI element grounding, and text localization, with strong performance in complex and cluttered scenes. Its core innovation, Parallel Box Decoding (PBD), predicts complete bounding box coordinates in a single parallel step rather than autoregressive token-by-token decoding, improving efficiency while preserving geometric consistency.…

Open weights other 3.8B parameters 32,768 tokens transformers

[View model](https://savrn.com/models/locateanything-3b)

Model · Image and text to text

### [blip2-flan-t5-xl](https://savrn.com/models/blip2-flan-t5-xl)

[Salesforce AI Research](https://savrn.com/model-publishers/salesforce)

BLIP-2 model, leveraging Flan T5-xl (a large language model). It was introduced in the paper BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models by Li et al. and first released in this repository. Disclaimer: The team releasing BLIP-2 did not write a model card for this model so this model card has been written by the Hugging Face team. BLIP-2 consists of 3 models: a CLIP-like image encoder, a Querying Transformer (Q-Former) and a large language model. The authors initialize the weights of the image encoder and large language model from pre-trained checkpoints and keep them frozen while training the Querying Transformer, which is a…

Open weights mit 3.9B parameters transformers

[View model](https://savrn.com/models/blip2-flan-t5-xl)

Model · Image and text to text

### [Rex-Omni](https://savrn.com/models/rex-omni)

[IDEA-Research](https://savrn.com/model-publishers/idea-research)

This model is Rex-Omni, a 3B-parameter Multimodal Large Language Model (MLLM) presented in the paper "Detect Anything via Next Point Prediction". It is compatible with the Hugging Face transformers library and is licensed under the IDEA License 1.0. src="https://img.shields.io/badge/RexOmni-Website-BADFDB?style=flat-square&logo=deno&logoColor=violet&color=BADFDB" alt="RexThinker Website" src="https://img.shields.io/badge/RexOmni-Paper-Red%25red?logo=arxiv&logoColor=red&color=yellow" alt="RexThinker Paper on arXiv" src="https://img.shields.io/badge/RexOmni-Weight-orange?logo=huggingface&logoColor=yellow" alt="RexThinker weight on Hugging Face"…

Open weights other 4.1B parameters 128,000 tokens transformers

[View model](https://savrn.com/models/rex-omni)

## Nanonets

[All models and datasets](https://savrn.com/model-publishers/nanonets)

## Versions

- [c3886ff00bb0](https://savrn.com/models/nanonets-ocr2-3b/versions/c3886ff00bb0) · current 2026-09-18

## Explore More

- [All image and text to text models](https://savrn.com/models/tasks/image-and-text-to-text)
- [Model comparisons](https://savrn.com/models/comparisons)
- [The model directory](https://savrn.com/models)
- [Open model prices by host](https://savrn.com/ai-index/pricing/open-models)

## Source

- Repository metadata, read 2026-10-20.
- [Hugging Face record](https://huggingface.co/nanonets/Nanonets-OCR2-3B)
- [How the hub is built](https://savrn.com/model-hub/methodology)