opendatalab/MinerU2.5-Pro-2604-1.2B
Primitive: /extract · Extract ·
qwen2_vl
MinerU2.5-Pro: Pushing the Limits of Data-Centric Document Parsing at Scale
MultimodalMultilingualEntities
Overview
Hardware: — drives latency, throughput & cost
| Size | 1.2B params |
|---|---|
| Tasks | /extract |
| License | apache-2.0 |
| Languages | zh, en |
| Latency | — |
| Throughput | — |
| Cost | — /1M tok |
Cost is approximate — computed from list GPU prices; your actual price depends on the provider you deploy SIE with.
Extraction
| Output kinds | Entities |
|---|---|
| Inputs | image |
| Max sequence length | — |
Benchmarks
olmOCR-Bench
Document text extraction accuracy across arxiv math, old scans, multi-column layouts, and tables
Corpus: 1,403 Queries: 1,403
Quality
accuracy 0.5910
Compare (0)Compare →