Gemini Flash vs docTR
Pick Gemini Flash for cheap bulk PDF to Markdown or prompt-driven JSON extraction, including mixed-language documents, when a hosted API is acceptable. Pick docTR when you need word-level text with coordinates, self-hosting under Apache-2.0, or your own trained models. docTR returns only words and lines, so it will not give you Markdown or tables.
Gemini Flashgoogle/gemini-flash | docTRmindee/doctr | |
|---|---|---|
| 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 | 0.9 s / page |
| Best for | Handwriting and mixed layouts | Word-level OCR with positions |
| Inputs | PDF, images | PDF, images |
| Languages | 100+ languages | Mainly Latin scripts |
| Max pages per request | 1,000 | 500 |
| Tables | ✓ Yes | No |
| Handwriting | ✓ Yes | No |
| Math | ✓ Yes | No |
| Bounding boxes | No | ✓ Yes |
| Confidence | No | ✓ Yes |
| 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, and docTR has no published benchmark.
- It can be prompted to return Markdown, JSON, HTML or plain text, while docTR has no Markdown or table output.
- The context window of up to 1M tokens fits long documents in one request.
Choose docTR if
- It exports a nested page, block, line and word structure with coordinates as JSON, while Gemini Flash gives no per-word confidence or coordinates by default.
- It is Apache-2.0 and runs locally on CPU or NVIDIA GPU, so it works where an air-gapped or on-prem deployment is required.
- There is no per-token bill, only your own compute, so cost does not vary with page density or output length.
Key differences
| Gemini Flash | docTR | |
|---|---|---|
| How you run it | Hosted API through the Gemini API or Vertex AI | Python library on PyTorch, with Docker images |
| Licence | Proprietary | Apache-2.0 |
| Price | Gemini 3 Flash Preview: $0.50 per 1M input tokens, $3.00 per 1M output tokens; Batch/Flex $0.25 input and $1.50 output | Free, self-hosted (you pay for compute) |
| Output formats | Markdown, JSON, HTML, plain text, shaped by your prompt with no fixed schema | JSON, plain text, bounding boxes |
| Tables and Markdown | Can be prompted to return Markdown | No table, formula or Markdown output; words and lines only |
| Benchmark | OmniDocBench v1.6 overall: 92.62 (Gemini 3 Flash) | Not published |
| Known failure | Cost per page depends on page density and output length; several newer Flash models rise from $0.75/$3.75 to $1.50/$7.50 per 1M tokens on January 1, 2027 | Cropping images gives poorer results (issue #2007); split-crop merging can misjoin repeated characters (issue #2229); ONNX export of MASTER truncates decoding steps (issue #2224) |
Sources
What each one is
Gemini Flash
General vision model with strong handwriting and layout understanding at a low price.
Gemini Flash detailsdocTR
Open-source OCR library from Mindee with separate text detection and recognition models.
docTR detailsTry both on your own file
Run the file once with each model and compare the output.
PDF, PNG, JPG, WebP, HEIC. Up to 20 MB each, 10 files.No file?
Use one, fall back to the other
{"model": "google/gemini-flash","fallbacks": ["mindee/doctr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Gemini Flash vs KrakenGemini Flash vs RapidOCRGPT vs docTRMistral OCR vs docTRdocTR vs DonutdocTR vs Phi-4-multimodalGemini Flash vs RolmOCRdocTR vs MonkeyOCR
Gemini Flash alternativesdocTR alternativesBest OCR models in 2026