HunyuanOCR 1.5 vs LightOnOCR-3-4B

HunyuanOCR 1.5 60

Memory 30, the only build from Tencent is the BF16 safetensors file at the root, 2.24 GB, and 2.24 GB × 1.2 = 2.7 GB, within 16 GB. ggml-org's GGUF (Q8_0 0.58 GB with a 0.73 GB projector) dates from 21 May 2026 and wasn't matched to version 1.5. Runtimes 15, Released Transformers 5.22.2 has the hunyuan_vl model, vLLM 0.33.0 registers HunYuanVLForConditionalGeneration, and llama.cpp b12242 has the architecture, which the card says the community build runs after conversion. Builds 0, Tencent publishes the BF16 weights only and tells users to convert to GGUF themselves. Terms 5, A custom licence that allows free commercial use under a use policy and below 100 million monthly active users, but not for anyone: it doesn't apply in the European Union, the United Kingdom or South Korea. Download 10, the Hub API reports gated false on 10 November 2026, so the files download with no account and nothing to accept. Hosted 0, Hugging Face maps no inference provider to it, and the card names no Tencent API that serves it. 30 + 15 + 0 + 5 + 10 + 0 = 60.

LightOnOCR-3-4B 70

Memory 30, the only build from LightOn is the BF16 safetensors file, 9.08 GB, and 9.08 GB × 1.2 = 10.9 GB, within 16 GB. No runtime library has a build of this model. Runtimes 15, the architecture is Qwen3_5ForConditionalGeneration, which vLLM 0.33.0, SGLang 0.5.23 and Transformers 5.22.2 run from the safetensors file, three at 5 each. Builds 0, LightOn publishes the BF16 weights only. Terms 15, Apache 2.0, OSI-approved, with nothing added, in the card's metadata and licence section. Download 10, the Hub API reports gated false on 10 November 2026, so the files download with no account and nothing to accept. Hosted 0, Hugging Face maps no inference provider to it, and the card names no hosted API for it. 30 + 15 + 0 + 15 + 10 + 0 = 70.

HunyuanOCR 1.5 45

File safety 25, safetensors only, no .bin, .pt or .pth files, and no Python at the root. Provenance 15, the files sit in the tencent organisation, which Tencent's GitHub repository for the model links to (10), and the Hub gives a SHA-256 for the large files (5). Safety evaluation 0, no safety results on the card. Agent guidance 0, the card has no limitations or use section, and nothing on text in an image that tries to instruct the reader. Licence risk 0, 15 less 5 for limits beyond the law (an acceptable use policy and three territories where the licence doesn't apply), 5 for terms the licensor can change (Tencent may update the policy from time to time) and 5 for the 100 million monthly active user threshold. Security programme 5, Tencent's Security Response Center takes vulnerability reports company-wide, as the Hy3 listing records (5). 25 + 15 + 0 + 0 + 0 + 5 = 45.

LightOnOCR-3-4B 55

File safety 25, one safetensors file, no .bin, .pt or .pth files and no Python in the repository. Provenance 15, the files sit in the lightonai organisation, which the Hub marks as verified and LightOn's GitHub repository for the models links to (10), and the Hub gives a SHA-256 for the large files (5). Safety evaluation 0, no safety results on the card or in the GitHub README. Agent guidance 0, Nothing on text in a document that tries to instruct the reader, or on safeguards to add. Licence risk 15, Apache 2.0 has no use policy, field-of-use limit, threshold or change clause, and nothing is added on the card. Security programme 0, no way to report a flaw found on the card or in the GitHub README. lighton.ai couldn't be read for a security.txt. 25 + 15 + 0 + 0 + 15 + 0 = 55.

HunyuanOCR 1.5 80

Chat template 20, chat_template.jinja in the repository (930 characters). Tool-calling format 25, Read for a document parser as what it takes and returns. Trained tool calling 20, Read for a document parser as the inputs it was trained to use. Structured output 5, Read for a document parser as typed outputs. Reasoning control 0, A document parser with no reasoning. Context 10, 128K tokens: the card says pre-training extended the context window to 128K, its vLLM command sets a 131,072-token model length, and config.json gives the language model 131,072 positions. 20 + 25 + 20 + 5 + 0 + 10 = 80.

LightOnOCR-3-4B 70

Chat template 20, chat_template.jinja in the repository (7,759 characters), Qwen 3.5's template with thinking off by default. Tool-calling format 25, Read for a document parser as what it takes and returns. Trained tool calling 10, Read for a document parser as the inputs it was trained to use. Structured output 15, Read for a document parser as typed outputs. Reasoning control 0, the card says it was trained and evaluated with thinking off, the template's default, and documents no thinking mode. Context 0, LightOn states no context length for it. 20 + 25 + 10 + 15 + 0 + 0 = 70.