Search public pages, research tools, and SAVRN solutions.
Open-weight model · Image to text
by IBM Granite ibm-granite/granite-vision-3.3-2b
Granite-vision-3.3-2b is a compact and efficient vision-language model, specifically designed for visual document understanding, enabling automated content extraction from tables, charts, infographics, plots, diagrams, and more.
What it takes to serve granite-vision-3.3-2b (3B 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 | 6.0 GB | 7.1 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 8-bit | 3.0 GB | 3.6 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x MI325X $2.00 |
| 4-bit | 1.5 GB | 1.8 GB | 1x MI300X (192 GB) Vultr |
$1.85 | 1x H100 $1.99 · 1x 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, read Sep 18, 2026.
By IBM Granite, published under apache-2.0, revision bf66a8244401.
Model Summary: Granite-vision-3.3-2b is a compact and efficient vision-language model, specifically designed for visual document understanding, enabling automated content extraction from tables, charts, infographics, plots, diagrams, and more. Granite-vision-3.3-2b introduces several novel experimental features such as image segmentation, doctags generation, and multi-page support (see Experimental Capabilities for more details) and offers enhanced safety when compared to earlier Granite vision models. The model was trained on a meticulously curated instruction-following data, comprising diverse public and synthetic datasets tailored to support a wide range of document understanding and general image tasks. Granite-vision-3.3-2b was trained by fine-tuning a Granite large language model with both image and text modalities.
Evaluations: We compare the performance of granite-vision-3.3-2b with previous versions of granite-vision models. Evaluations were done using the standard llms-eval benchmark and spanned multiple public benchmarks, with particular emphasis on document understanding tasks while also including general visual question-answering benchmarks.
17 files, 6.0 GB in total. The weights are 2 files totalling 6.0 GB in safetensors.
| File | Type | Size | SHA-256 |
|---|---|---|---|
| model-00001-of-00002.safetensors | Weights | 5.0 GB | c4d07df4cdb4 |
| model-00002-of-00002.safetensors | Weights | 952.2 MB | 68c1ad6cbd0d |
| added_tokens.json | Configuration | 107 B | — |
| chat_template.json | Configuration | 1.6 KB | — |
| config.json | Configuration | 2.8 KB | — |
| generation_config.json | Configuration | 132 B | — |
| model.safetensors.index.json | Configuration | 83.1 KB | — |
| preprocessor_config.json | Configuration | 1.6 KB | — |
| processor_config.json | Configuration | 174 B | — |
| special_tokens_map.json | Configuration | 701 B | — |
| README.md | Documentation | 11.9 KB | — |
| example.png | Other | 433.7 KB | 6df7c7665c5d |
| .gitattributes | Repository | 1.6 KB | — |
| merges.txt | Tokenizer | 441.8 KB | — |
| tokenizer.json | Tokenizer | 3.5 MB | — |
| tokenizer_config.json | Tokenizer | 6.0 KB | — |
| vocab.json | Tokenizer | 777.0 KB | — |
Released by IBM Granite through its official repository on Hugging Face. Read the license.
| Precision | Weights in memory |
|---|---|
| As published | 6.0 GB |
| 16-bit | 6.0 GB |
| 8-bit | 3.0 GB |
| 4-bit | 1.5 GB |
Weights only, from the published parameter count; the key-value cache and runtime add to this.
About 7.1 GB at 16-bit and 1.8 GB at 4-bit: the weights (3B parameters) plus a working margin. A long context needs more.
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.
Yes. granite-vision-3.3-2b is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.
131,072 tokens, from the maximum position embeddings in its published configuration.
This Hub repository contains a HuggingFace's transformers implementation of the original Kosmos-2 model from Microsoft. Use the code below to get started with the model. This model is capable of performing different tasks through changing the prompts. First, let's define a function to run a prompt. Here are the tasks Kosmos-2 could perform: Once you have the entities, you can use the following helper function to draw their bounding bboxes on the image
NuExtract3 is a unified 4B vision-language reasoning model for document understanding. It combines strong structured information extraction with high-quality image-to-Markdown conversion, making it suitable for extraction pipelines, OCR, and RAG preprocessing for all types of documents such as scans, receipts, forms, invoices, contracts or tables. Try it out in the space! - Multilingual documents. - Reasoning and non-reasoning inference modes. - Template generation for structured extraction from natural language or input document. We benchmarked NuExtract on NuMind's internal structured benchmark, measuring model's performances on ~600 documents of diverse types including invoices, movie…
Join our WeChat and Discord community Use GLM-OCR's API GLM-OCR is a multimodal OCR model for complex document understanding, built on the GLM-V encoder–decoder architecture. It introduces Multi-Token Prediction (MTP) loss and stable full-task reinforcement learning to improve training efficiency, recognition accuracy, and generalization. The model integrates the CogViT visual encoder pre-trained on large-scale image–text data, a lightweight cross-modal connector with efficient token downsampling, and a GLM-0.5B language decoder. Combined with a two-stage pipeline of layout analysis and parallel recognition based on PP-DocLayout-V3, GLM-OCR delivers robust and high-quality OCR performance…
Join our WeChat and Discord community Use GLM-OCR's API GLM-OCR is a multimodal OCR model for complex document understanding, built on the GLM-V encoder–decoder architecture. It introduces Multi-Token Prediction (MTP) loss and stable full-task reinforcement learning to improve training efficiency, recognition accuracy, and generalization. The model integrates the CogViT visual encoder pre-trained on large-scale image–text data, a lightweight cross-modal connector with efficient token downsampling, and a GLM-0.5B language decoder. Combined with a two-stage pipeline of layout analysis and parallel recognition based on PP-DocLayout-V3, GLM-OCR delivers robust and high-quality OCR performance…
Full BF16 version of the model. We recommend this variant for inference and further fine-tuning. LightOnOCR-1B is a compact, end-to-end vision–language model for Optical Character Recognition (OCR) and document understanding. It achieves state-of-the-art accuracy in its weight class while being several times faster and cheaper than larger general-purpose VLMs. Highlights LightOnOCR combines a Vision Transformer encoder(Pixtral-based) with a lightweight text decoder(Qwen3-based) distilled from high-quality open VLMs. It is optimized for document parsing tasks, producing accurate, layout-aware text extraction from high-resolution pages. All benchmarks evaluated using vLLM on the Olmo-Bench.…
This model was converted to MLX format from zai-org/GLM-OCR using mlx-vlm version 0.3.10. Refer to the original model card for more details on the model.
