Gemini Flash vs Florence-2
Pick Gemini Flash for bulk PDF to Markdown or JSON, mixed-language files and long documents, and accept that you pay per token on a hosted API. Pick Florence-2 for English scene text and short labels that need bounding boxes, on your own hardware under an MIT licence. Florence-2 cannot produce Markdown or tables and fails on CJK and handwriting.
Gemini Flashgoogle/gemini-flash | Florence-2microsoft/florence-2 | |
|---|---|---|
| Type | Vision model | Open source |
| Price | 15 credits / page | 1 credit / page |
| Per 1,000 pages | $15.00 | $1.00 |
| Typical speed | 1.6 s / page | 0.8 s / page |
| Best for | Handwriting and mixed layouts | Lightweight OCR with boxes |
| Inputs | PDF, images | images |
| Languages | 100+ languages | English |
| Max pages per request | 1,000 | 500 |
| Tables | ✓ Yes | No |
| Handwriting | ✓ Yes | No |
| Math | ✓ Yes | No |
| Bounding boxes | No | ✓ Yes |
| Confidence | No | No |
| Forms | No | No |
| Multilingual | ✓ Yes | No |
| 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
- Gemini 3 Flash scores 92.62 on OmniDocBench v1.6, close to Gemini 3 Pro at 92.91.
- It can be prompted to return Markdown, JSON or HTML, so full-page documents and tables convert in one pass.
- The context window of up to 1M tokens fits long documents in a single request.
Choose Florence-2 if
- It is MIT licensed and self-hosted, so you pay only for compute and can run it on your own infrastructure.
- OCR_WITH_REGION returns quadrilateral boxes per text line, which suits scene text and short labels.
- At 0.23B (base) and 0.77B (large) parameters it is small enough for lightweight edge OCR in English.
Key differences
| Gemini Flash | Florence-2 | |
|---|---|---|
| How you run it | Hosted API (Gemini API or Vertex AI) | Self-hosted via Transformers |
| Licence | Proprietary | MIT |
| Price | Gemini 3 Flash Preview: $0.50 per 1M input tokens, $3.00 per 1M output tokens. Batch/Flex: $0.25 input, $1.50 output. | Free, you pay for compute |
| Output formats | Markdown, JSON, HTML, plain text, driven by the prompt | Plain text and bounding boxes only. No Markdown, tables or LaTeX. |
| Model size | Not published | 0.23B (base), 0.77B (large) |
| Benchmark | OmniDocBench v1.6 overall: 92.62 (Gemini 3 Flash, independent) | TextCaps CIDEr zero-shot 72.8 and COCO detection mAP 37.5 (large, from the authors). No document benchmark published. |
| Known failure | Cost per page varies with page density and output length. No fixed schema or confidence scores. | Spelling errors and missing spaces on cursive handwriting and French text. Chinese output is mostly wrong because the vocabulary lacks CJK characters. |
Sources
- Gemini API pricing
- OmniDocBench leaderboard
- Florence-2-large model card
- HF blog OCR test
- HF discussion Chinese very bad
What each one is
Gemini Flash
General vision model with strong handwriting and layout understanding at a low price.
Gemini Flash detailsFlorence-2
Small Microsoft vision model (0.23B and 0.77B) with OCR and OCR-with-regions tasks that return text with boxes.
Florence-2 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": ["microsoft/florence-2"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Gemini Flash alternativesFlorence-2 alternativesBest OCR models in 2026