LlamaParse vs Chandra
Pick Chandra if you need multilingual scans, tables, forms or math, or want to self-host on your own GPU. Pick LlamaParse if you want a hosted API for mixed PDF and office files inside a LlamaIndex RAG pipeline and prefer to pay per page with a free monthly allowance. Validate LlamaParse output on bank statements and poor scans, and validate Chandra output on old or low-resource-script scans.
LlamaParsellamaindex/llamaparse | Chandradatalab/chandra | |
|---|---|---|
| Type | OCR API | Open source |
| Price | 5 credits / page | 2 credits / page |
| Per 1,000 pages | $5.00 | $2.00 |
| Typical speed | 3.0 s / page | 2.0 s / page |
| Best for | RAG ingestion | Complex layouts and handwriting |
| Inputs | PDF, images, DOCX | PDF, images |
| Languages | 80+ languages | 40+ languages |
| Max pages per request | 1,000 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | No | ✓ Yes |
| Math | ✓ Yes | ✓ Yes |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | No | No |
| Forms | No | ✓ Yes |
| 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 LlamaParse if
- It handles PDFs and office files in one hosted service, so you do not run any model yourself.
- It fits RAG ingestion inside LlamaIndex pipelines, and re-parsing the same file within 48 hours is free because results are cached.
- It has a wide price range, from $1.25 per 1,000 pages in Fast mode to agentic modes for hard layouts, with 10,000 free credits per month.
Choose Chandra if
- You can self-host the open weights on a single NVIDIA GPU with vLLM, or use the hosted Datalab API if you prefer.
- It scores 85.8 on olmOCR-bench and 77.8% averaged across 43 languages, so it suits multilingual scanned documents.
- It returns Markdown, HTML and JSON with bounding boxes and labels per block, which helps with tables, forms and math.
Key differences
| LlamaParse | Chandra | |
|---|---|---|
| How you run it | Hosted API only | Self-host the open weights (vLLM or Hugging Face) or use the hosted Datalab API |
| Licence | Proprietary | Code Apache-2.0; weights modified OpenRAIL-M, free for research, personal use and companies under $2M funding/revenue |
| Price | $1.25 per 1,000 credits. Fast 1 credit/page, Cost Effective 3, Agentic 10, Agentic Plus 45. 10K free credits per month | Self-hosted: free within license terms, you pay for compute. Datalab API: $1, $3 or $7.50 per 1,000 pages; $20/month free tier with a work email |
| Output formats | Markdown, JSON, plain text, images | Markdown, HTML, JSON, bounding boxes |
| Benchmark | Not published | olmOCR-bench 85.8 overall (Chandra 2, authors' figure) |
| Known failure | Invented transactions on bank statements in Accurate mode; hallucinations on low-quality scans with no confidence score or error | Token-loop hallucinations on some Russian/Kazakh scans; weak on Urdu (43.2%), Sanskrit (51.1%) and old scans (51.1 on olmOCR-bench) |
| Speed | Premium mode took 4 to 5 minutes in one report; one job stuck for days | About 1.44 pages/s on one H100 with vLLM; one reviewer measured 66 seconds per page on an RTX 4090 with the Hugging Face path |
Sources
- LlamaIndex pricing
- GitHub issue #450
- GitHub issue #483
- Chandra GitHub README
- Datalab pricing
- GitHub issue #51 token loops
What each one is
Chandra
Datalab's OCR model that keeps layout and handles forms, tables and handwriting.
Chandra 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": "llamaindex/llamaparse","fallbacks": ["datalab/chandra"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
LlamaParse vs SuryaLlamaParse vs olmOCRMathpix Convert vs ChandraClaude vs ChandraLlamaParse vs Reducto ParseLlamaParse vs DoclingLlamaParse vs Adobe Acrobat OCRChandra vs Pixtral
LlamaParse alternativesChandra alternativesBest OCR models in 2026