We also ship smaller models for line-level text detection and ocr error detection. It works on a range of documents (see usage and benchmarks).
Alternatives to surya-ocr-2.5
120 open-weight models that do the same job as surya-ocr-2.5 (689M OCR), ranked by how close they are in task and size, then by popularity and how recent they are. Licence and VRAM are shown for each.
d1-omni-600M is a 600M parameter decision model built on LFM2.5-Encoder-350M. You give it a state (text or JSON, with images or a voice clip) and a set of named questions.
587M125K ctx lfm1.07.7K 167Oct 2026- Related category
- Similar size (587M vs 689M)
T5Gemma is a family of lightweight yet powerful encoder-decoder research models from Google, built by adapting pretrained decoder-only Gemma models into encoder-decoder ones.
786M32K ctx Gemma Terms of Use32.2K 215Oct 2025- Related category
- Similar size (786M vs 689M)
Intern-Decision-0.8B is a multimodal structured decision model fine-tuned from Qwen3.5-0.8B. It accepts a shared state, a schema of named questions, and optional images, and returns an answer distribution for every question in one model forward pass.
853M256K ctx Apache 2.04.1K 41Sept 2026- Related category
- Similar size (853M vs 689M)
- Commercial use OK
NVIDIA Nemotron Parse v1.1 is designed to understand document semantics and extract text and tables elements with spatial grounding.
957M9K ctx NVIDIA Open Model License110.8K 174Nov 2025- Related category
- Similar size (957M vs 689M)
- Commercial use OK
FastVLM was introduced in FastVLM: Efficient Vision Encoding for Vision Language Models. (CVPR 2025)
759M32K ctx Apple Machine Learning Research License22.4K 410Aug 2025- Related category
- Similar size (759M vs 689M)
MiniCPM-V 4.7 is the latest model in the MiniCPM-V series, available in a 1B variant. It builds on the family's single-image, multi-image, and video understanding capabilities, combining stronger foundation capabilities with efficient visual encoding and on-device inference.
1.6B256K ctx Apache 2.0468.8K 418Oct 2026- Related category
- Larger (1.6B vs 689M)
- Commercial use OK
LFM2.5‑VL-450M is Liquid AI's refreshed version of the first vision-language model, LFM2-VL-450M, built on an updated backbone LFM2.5-350M and tuned for stronger real-world performance. Find more about LFM2.5 family of models in our blog post.
449M125K ctx lfm1.041.4K 240Apr 2026- Related category
- Similar size (449M vs 689M)
A Pocket-Sized MLLM for Ultra-Efficient Image and Video Understanding on Your Phone
1.3B256K ctx Apache 2.0194.5K 1.3KApr 2026- Related category
- Similar size (1.3B vs 689M)
- Commercial use OK
NVIDIA Nemotron Parse v1.2 is designed to understand document semantics and extract text and tables elements with spatial grounding.
936M9K ctx NVIDIA Open Model License10.2K 83Feb 2026- Related category
- Similar size (936M vs 689M)
- Commercial use OK
We introduce InternVL3, an advanced multimodal large language model (MLLM) series that demonstrates superior overall performance.
938M32K ctx Qwen License72.4K 11Apr 2025- Related category
- Similar size (938M vs 689M)
We introduce InternVL3, an advanced multimodal large language model (MLLM) series that demonstrates superior overall performance.
938M32K ctx Apache 2.054.6K 85Apr 2025- Related category
- Similar size (938M vs 689M)
- Commercial use OK
Check out also the Google Colab demo to run Llava on a free-tier Google Colab instance:
894MApache 2.0187.7K 61Aug 2024- Related category
- Similar size (894M vs 689M)
- Commercial use OK
Dripper(MinerU-HTML) is an advanced HTML main content extraction tool based on Large Language Models (LLMs). It provides a complete pipeline for extracting primary content from HTML pages using LLM-based classification and state machine-guided generation.
752M40K ctx Apache 2.0687 62Nov 2025- Related category
- Similar size (752M vs 689M)
- Commercial use OK
A Pocket-Sized MLLM for Ultra-Efficient Image and Video Understanding on Your Phone
1.3B256K ctx Apache 2.027.1K 34May 2026- Related category
- Similar size (1.3B vs 689M)
- Commercial use OK
🎉 A live demo is available at hugging-apps/hpd-parsing. Thanks to multimodalart for building it! 🚀 We release HPD-Parsing, a 1B hierarchical parallel document parser. It achieves 94.91% overall on OmniDocBench v1.6 and 4,752 TPS peak throughput.
1.1B40K ctx Apache 2.02.3K 102Jul 2026- Related category
- Similar size (1.1B vs 689M)
- Commercial use OK
The H2OVL-Mississippi-800M is a compact yet powerful vision-language model from H2O.ai, featuring 0.8 billion parameters. Despite its small size, it delivers state-of-the-art performance in text recognition, excelling in the Text Recognition segment of OCRBench and outperforming much larger models…
826M8K ctx Apache 2.027.3K 43Oct 2024- Related category
- Similar size (826M vs 689M)
- Commercial use OK
Over recent months, we have intensified our focus on developing foundation models that deliver exceptional utility and performance.
2.3B256K ctx Apache 2.05.4M 498Feb 2026- Related category
- Larger (2.3B vs 689M)
- Commercial use OK
- Widely used
LFM2‑VL is Liquid AI's first series of multimodal models, designed to process text and images with variable resolutions. Built on the LFM2 backbone, it is optimized for low-latency and edge AI applications.
451M125K ctx lfm1.018.7K 152Aug 2025- Related category
- Similar size (451M vs 689M)
LFM2.5-VL-450M-Extract extracts user-defined fields from images and returns them as JSON. It is Liquid AI's first vision model in the Liquid Nanos collection—compact, task-specific models built for production workflows—and extends the Extract family alongside LFM2-350M-Extract for text documents.
449M125K ctx lfm1.02.1K 57May 2026- Related category
- Similar size (449M vs 689M)
SingGuard-NSFA is a dual-mode guardrail framework for securing agentic AI systems against operational threats such as prompt injection, sensitive information extraction, malicious code requests, dangerous tool misuse, and resource exhaustion.
1.1B256K ctx Apache 2.01.8K 10Jul 2026- Related category
- Similar size (1.1B vs 689M)
- Commercial use OK
We introduce InternVL3.5, a new family of open-source multimodal models that significantly advances versatility, reasoning capability, and inference efficiency along the InternVL series.
1.1B40K ctx Apache 2.012.9K 31Aug 2025- Related category
- Similar size (1.1B vs 689M)
- Commercial use OK
Model type: LLaVA Interleave is an open-source chatbot trained by fine-tuning LLM on multimodal instruction-following data. It is an auto-regressive language model, based on the transformer architecture. Base LLM: Qwen/Qwen1.5-7B-Chat
864M32K ctx Qwen Research License17.9K 37Jul 2024- Related category
- Similar size (864M vs 689M)
LFM3-VL-2B is a vision-language model for on-device visual understanding. It takes text, images, and video as input, and combines the LFM3 language model with a new in-house vision encoder.
1.9B125K ctx lfm1.030.7K 288Oct 2026- Related category
- Larger (1.9B vs 689M)
Meet Qwen3-VL — the most powerful vision-language model in the Qwen series to date.
2.1B256K ctx Apache 2.01.2M 522Oct 2025- Related category
- Larger (2.1B vs 689M)
- Commercial use OK