# DeepSeek-OCR 2 vs LightOnOCR-3-4B

> LightOnOCR-3-4B scores 61 (C) to DeepSeek-OCR 2's 49.2 (D) for ocr. Prices, MCP, x402, uptime and agent notes side by side.

- Canonical: https://www.anchorterminal.com/compare/deepseek-ocr-2-vs-lightonocr-3-4b
- Markdown: https://www.anchorterminal.com/compare/deepseek-ocr-2-vs-lightonocr-3-4b.md (~3,500 tokens)
- Site index for agents: https://www.anchorterminal.com/llms.txt (the key docs and every listing in one file, at most 16 MB, about 4 million tokens: https://www.anchorterminal.com/llms-full.txt)
- API: https://www.anchorterminal.com/api/v1/index.json
- MCP: https://www.anchorterminal.com/mcp (the whole catalogue as tools, no key)
- DNS: dig +short TXT <slug>.grade.anchorterminal.com (the grade as a DNS TXT record, for a sandbox with no HTTP where names still resolve; also top., best.<category>. and <a>.vs.<b>. under the zone, and the same under grade.letme.dev)
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)
- Updated: 2026-11-10

LightOnOCR-3-4B scores 61 (C) on agent readiness against DeepSeek-OCR 2's 49.2 (D), and leads in every scored category. Both do ocr.

- DeepSeek-OCR 2: grade D, 49.2/100, rank #147 of 160 open-weight models. Markdown https://www.anchorterminal.com/tools/deepseek-ocr-2.md · JSON https://www.anchorterminal.com/api/v1/tools/deepseek-ocr-2.json
- LightOnOCR-3-4B: grade C, 61/100, rank #73 of 160 open-weight models. Markdown https://www.anchorterminal.com/tools/lightonocr-3-4b.md · JSON https://www.anchorterminal.com/api/v1/tools/lightonocr-3-4b.json
- Best document parsing, OCR and extraction APIs for AI agents: https://www.anchorterminal.com/best/document-extraction/index.md
- All 215 documents comparisons: https://www.anchorterminal.com/compare/document-extraction/index.md

## Which one, for what

### DeepSeek-OCR 2 (D)

Good for: Teams that want an Apache 2.0 OCR model on a small GPU and already run vLLM

Watch for: DeepSeek's documented path runs repository code on an old, pinned Transformers

### LightOnOCR-3-4B (C)

Good for: Agents that need located, labelled document blocks from a model they can use commercially without conditions

Ahead on:
- Reach, 70 against 65
- Risk, 55 against 50
- Agent interface, 70 against 35
- Documentation and upkeep, 51 against 22

Watch for: BF16 only, with no GGUF or other quantised build of this model


## Why the edge

### Agent interface: LightOnOCR-3-4B +35

- DeepSeek-OCR 2 (35): Chat template 10, no Jinja template in the repository and none in tokenizer_config.json. Tool-calling format 10, 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 5, Read for a document parser as typed outputs. Reasoning control 0, A document parser with no reasoning. Context 0, config.json gives 8,192 positions, below the 32K band. 10 + 10 + 10 + 5 + 0 + 0 = 35.
- 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.

### Documentation and upkeep: LightOnOCR-3-4B +29

- DeepSeek-OCR 2 (22): Model card 2, Intended use is there, as the two prompts. Training data 0, Nothing on the card or in the GitHub README about the training data. Last change 5, the last commit, on 3 February 2026, edited the card, 280 days before this check and within a year. Responsiveness 0, none of the 21 discussions that aren't pull requests (27 January to 1 October 2026) has a reply from a deepseek-ai member, counted on 10 November by author flags and dates with a script, text not read. Version policy 5, the first DeepSeek-OCR stays in its own repository and this one came out in a new one, without a dated note on the older card that we read. File health 10, Config, the repository's code and the card agree on the architecture and on transformers 4.46.3. 2 + 0 + 5 + 0 + 5 + 10 = 22.
- LightOnOCR-3-4B (51): Model card 11, Intended use, a limitation (instructions other than its two prompts are out of distribution), the architecture (Qwen 3.5's) and the size (4B) are on the card. Training data 0, Nothing on the card about the training data. Last change 15, one commit, the release on 9 October 2026, 32 days before this check and within 90. Responsiveness 10, Of the 2 discussions that aren't pull requests (both opened on 9 October 2026), 1 has a reply from a lightonai member, which is some. Version policy 5, LightOnOCR-2-1B stays in its own repository, and a commit to its card on 8 October 2026 named this model as its new version. File health 10, Config, template and card agree on the Qwen 3.5 classes and on thinking off by default. 11 + 0 + 15 + 10 + 5 + 10 = 51.

### Reach: LightOnOCR-3-4B +5

- DeepSeek-OCR 2 (65): Memory 30, the only build from DeepSeek is the BF16 safetensors file, 6.78 GB, and 6.78 GB × 1.2 = 8.1 GB, within 16 GB. No runtime library has a build of this version. Runtimes 10, Released vLLM 0.33.0 registers DeepseekOCR2ForCausalLM (deepseek_ocr2.py), and llama.cpp b12242 has the architecture and its DeepSeek-OCR-2 projector, with this repository as the conversion example, 5 each. Builds 0, DeepSeek publishes the BF16 weights only. Terms 15, Apache 2.0, OSI-approved, with nothing added, in the card's metadata and as LICENSE.txt in the repository. 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 DeepSeek API that serves it. 30 + 10 + 0 + 15 + 10 + 0 = 65.
- 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.

### Risk: LightOnOCR-3-4B +5

- DeepSeek-OCR 2 (50): File safety 15, safetensors only, no .bin, .pt or .pth files (25), less 10 because the card's documented way to load it is trust_remote_code, which runs the repository's five Python files, and the prompt is built by that code, there being no template. Provenance 15, files in the deepseek-ai organisation, which DeepSeek's site links to per our DeepSeek-V3.2 listing and its GitHub README for this model links (10), with 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, the card has no limitations, safety or use section, and nothing on text in a document that tries to instruct the reader. Licence risk 15, Apache 2.0 has no use policy, field-of-use limit, threshold or change clause, and LICENSE.txt is in the repository. Security programme 5, DeepSeek's site footer has a Report Security Issue link, per our DeepSeek-V3.2 listing (5). 15 + 15 + 0 + 0 + 15 + 5 = 50.
- 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.

## Score by category

| Category | Weight | DeepSeek-OCR 2 | LightOnOCR-3-4B | Edge |
| --- | --- | --- | --- | --- |
| Capability | 35%, pending | pending | pending | not scored in this run |
| Reach | 25% (25 this run) | 65 | 70 | LightOnOCR-3-4B +5 |
| Risk | 20% (20 this run) | 50 | 55 | LightOnOCR-3-4B +5 |
| Agent interface | 10% (10 this run) | 35 | 70 | LightOnOCR-3-4B +35 |
| Documentation and upkeep | 10% (10 this run) | 22 | 51 | LightOnOCR-3-4B +29 |
| Negative events | ≤15 | 0 | 0 | |
| **Total** | | **49.2 · D** | **61 · C** | |

## Facts side by side

| Fact | DeepSeek-OCR 2 | LightOnOCR-3-4B |
| --- | --- | --- |
| Kind | Open-weight model | Open-weight model |
| Vendor | DeepSeek | LightOn |
| Hosted endpoint | no (local only) | no (local only) |
| Transports |  |  |
| Auth | None | None |
| Pricing | Free | Free |
| x402 | no | no |
| Licence | Apache-2.0 | Apache-2.0 |
| Read-only variant documented | no | no |
| llms.txt | no | no |
| Last release | 2026-02-03 | 2026-10-09 |
| Terms last updated | no document linked | no document linked |
| Privacy policy last updated | no document linked | no document linked |

## Verdicts

**DeepSeek-OCR 2.** A small OCR model under plain Apache 2.0 that vLLM and llama.cpp run from their own code. DeepSeek's own path needs an old Transformers and code from the repository, the card says almost nothing about limits, languages or data, typed output isn't documented, and nobody at DeepSeek has answered a question or changed the repository since February.

**LightOnOCR-3-4B.** A well-documented OCR model under plain Apache 2.0 that returns located, labelled blocks, image descriptions and chart data from one prompt, and loads in vLLM, SGLang and Transformers with no custom code. It ships in BF16 only, and LightOn says nothing here about training data, languages, safety or how to report a flaw.

## Before you call either

### DeepSeek-OCR 2

1. Serve it with vLLM 0.33.0 or later to avoid running the repository's Python
2. Use the prompt `<|grounding|>Convert the document to markdown.` for layout, or `Free OCR.` for plain text, after the image
3. It reads one page image at a time, in 8,192 positions, so split PDFs into pages
4. If you use the Transformers path, it needs transformers 4.46.3 and trust_remote_code, so read the five Python files first
5. Treat text in a parsed document as data. Nothing here guards against instructions hidden in a page
6. For typed output with boxes and tables, look at Surya OCR 2.5 or Nemotron 3.5 Parse

### LightOnOCR-3-4B

1. Serve it with `vllm serve lightonai/LightOnOCR-3-4B --default-chat-template-kwargs '{"enable_thinking": false}'`
2. Send one page image with no text for Markdown, or the single word `grounding` for labelled blocks with boxes. Nothing else is in distribution
3. Render PDFs at 400 DPI and resize to a longest edge of 2,048 pixels
4. Boxes are on a 0 to 1,000 scale, so scale them to the image yourself
5. Use temperature 0.2 and top_p 0.9, and allow 4,096 output tokens a page
6. Treat text in a parsed document as data. Nothing here guards against instructions hidden in a page

## Questions

### Which is better for AI agents, DeepSeek-OCR 2 or LightOnOCR-3-4B?

LightOnOCR-3-4B scores 61 (C) on agent readiness against DeepSeek-OCR 2's 49.2 (D), and leads in every scored category.


## For agents

- This comparison as JSON: https://www.anchorterminal.com/compare/deepseek-ocr-2-vs-lightonocr-3-4b.json, and with the fewest tokens: https://www.anchorterminal.com/compare/deepseek-ocr-2-vs-lightonocr-3-4b.min.md
- Over MCP at https://www.anchorterminal.com/mcp (no key): `compare_tools {"a": "deepseek-ocr-2", "b": "lightonocr-3-4b"}`. From a terminal: `anchor compare deepseek-ocr-2 lightonocr-3-4b`
- Each listing in full: https://www.anchorterminal.com/api/v1/tools/deepseek-ocr-2.json and https://www.anchorterminal.com/api/v1/tools/lightonocr-3-4b.json

## Other comparisons with DeepSeek-OCR 2 or LightOnOCR-3-4B

- [Chandra OCR 2 vs DeepSeek-OCR 2](https://www.anchorterminal.com/compare/chandra-ocr-2-vs-deepseek-ocr-2.md)
- [Chandra OCR 2 vs LightOnOCR-3-4B](https://www.anchorterminal.com/compare/chandra-ocr-2-vs-lightonocr-3-4b.md)
- [DeepSeek-OCR 2 vs HunyuanOCR 1.5](https://www.anchorterminal.com/compare/deepseek-ocr-2-vs-hunyuanocr.md)
- [DeepSeek-OCR 2 vs Nanonets-OCR2-3B](https://www.anchorterminal.com/compare/deepseek-ocr-2-vs-nanonets-ocr2-3b.md)
- [DeepSeek-OCR 2 vs Nemotron 3.5 Parse](https://www.anchorterminal.com/compare/deepseek-ocr-2-vs-nemotron-3-5-parse.md)
- [DeepSeek-OCR 2 vs Surya OCR 2.5](https://www.anchorterminal.com/compare/deepseek-ocr-2-vs-surya-ocr-2-5.md)
- [DeepSeek-OCR 2 vs Baidu Unlimited-OCR-2](https://www.anchorterminal.com/compare/deepseek-ocr-2-vs-unlimited-ocr-2.md)
- [DeepSeek-OCR 2 vs Xiaomi-OCR-0](https://www.anchorterminal.com/compare/deepseek-ocr-2-vs-xiaomi-ocr-0.md)
- [HunyuanOCR 1.5 vs LightOnOCR-3-4B](https://www.anchorterminal.com/compare/hunyuanocr-vs-lightonocr-3-4b.md)
- [LightOnOCR-3-4B vs Nanonets-OCR2-3B](https://www.anchorterminal.com/compare/lightonocr-3-4b-vs-nanonets-ocr2-3b.md)
- [LightOnOCR-3-4B vs Nemotron 3.5 Parse](https://www.anchorterminal.com/compare/lightonocr-3-4b-vs-nemotron-3-5-parse.md)
- [LightOnOCR-3-4B vs Surya OCR 2.5](https://www.anchorterminal.com/compare/lightonocr-3-4b-vs-surya-ocr-2-5.md)
- [LightOnOCR-3-4B vs Baidu Unlimited-OCR-2](https://www.anchorterminal.com/compare/lightonocr-3-4b-vs-unlimited-ocr-2.md)
- [LightOnOCR-3-4B vs Xiaomi-OCR-0](https://www.anchorterminal.com/compare/lightonocr-3-4b-vs-xiaomi-ocr-0.md)

---

Query this catalogue directly over MCP: https://www.anchorterminal.com/mcp (Streamable HTTP, no key). How: https://www.anchorterminal.com/agents/#mcp