Gemini Flash vs Unlimited OCR
Pick Gemini Flash for cheap bulk PDF to Markdown or JSON extraction through a hosted API, especially with mixed-language documents. Pick Unlimited OCR when you need MIT-licensed self-hosting, bounding boxes, or long reports and books parsed in one pass on your own NVIDIA GPU. The main split is a token-billed proprietary API versus a free model you run yourself.
Gemini Flashgoogle/gemini-flash | Unlimited OCRbaidu/unlimited-ocr | |
|---|---|---|
| Type | Vision model | Open source |
| Price | 15 credits / page | 2 credits / page |
| Per 1,000 pages | $15.00 | $2.00 |
| Typical speed | 1.6 s / page | 1.2 s / page |
| Best for | Handwriting and mixed layouts | Books and very long documents |
| Inputs | PDF, images | PDF, images |
| Languages | 100+ languages | 30+ languages |
| Max pages per request | 1,000 | 1,000 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | ✓ Yes | No |
| Math | ✓ Yes | ✓ Yes |
| Bounding boxes | No | ✓ Yes |
| Confidence | No | No |
| Forms | No | No |
| Multilingual | ✓ Yes | ✓ Yes |
| On OCRRouter | Available | Available |
Prices are what you pay on OCRRouter. Speed is the provider's typical figure, not a benchmark we ran.
Which to choose
Choose Gemini Flash if
- It is a hosted API with no GPU to run, and Gemini 3 Flash Preview costs $0.50 per 1M input tokens and $3.00 per 1M output tokens ($0.25 and $1.50 with Batch/Flex).
- Gemini 3 Flash scores 92.62 on OmniDocBench v1.6, close to Gemini 3 Pro at 92.91.
- It can return Markdown, JSON, HTML or plain text from a prompt or schema, and its 1M-token context fits long documents in one request.
Choose Unlimited OCR if
- It is MIT licensed, so commercial use is allowed with no copyleft, and you can self-host it with vLLM, SGLang or transformers.
- It emits block types and bounding boxes alongside text, which suits layout-aware extraction.
- It parses 40+ pages in a single forward pass with roughly constant memory, and Unlimited OCR 2 scores 77.1 on MarathonDocBench versus 33.7 for the next model, per Baidu.
Key differences
| Gemini Flash | Unlimited OCR | |
|---|---|---|
| Licence | Proprietary | MIT |
| How you run it | Hosted API (Gemini API or Vertex AI) only; not for air-gapped or on-prem use | Self-host with vLLM, SGLang or transformers, or use the hosted endpoint in Baidu Cloud's OCR product; needs an NVIDIA GPU with CUDA 12.9 or 13.0 |
| Price | $0.50 input / $3.00 output per 1M tokens (Gemini 3 Flash Preview); cost per page varies with page density and output length | Free to self-host, you pay for compute; hosted price Not published |
| Output formats | Markdown, JSON, HTML, plain text; prompt-driven with no fixed schema or confidence scores | Markdown, plain text, bounding boxes |
| Benchmark | 92.62 on OmniDocBench v1.6 (Gemini 3 Flash) | 46.17 mean on ParseBench; 84.6 on MPDocBench and 77.1 on MarathonDocBench (Unlimited OCR 2, author-reported) |
| Long documents | Context window up to 1M tokens | 40+ pages in one pass, but context is capped at 32,768 tokens and multi-page input only works in base mode |
| Known failure | Several newer Flash models move from $0.75/$3.75 to $1.50/$7.50 per 1M tokens on January 1, 2027 | Text repetition and looping (issue #55), hallucinated text (issue #58), and styling mostly lost (0.97 text formatting score on ParseBench) |
Sources
- Gemini API pricing
- OmniDocBench leaderboard
- Hugging Face model card
- MarkTechPost on Unlimited OCR 2
- Issue #55 repetition
What each one is
Gemini Flash
General vision model with strong handwriting and layout understanding at a low price.
Gemini Flash detailsUnlimited OCR
Open-source model from Baidu that reads hundreds of pages in one pass with flat memory use.
Unlimited OCR detailsTry both on your own file
Run the file once with each model and compare the output.
Use one, fall back to the other
{"model": "google/gemini-flash","fallbacks": ["baidu/unlimited-ocr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Gemini Flash alternativesUnlimited OCR alternativesBest OCR models in 2026