Model card
Chandra v0.1.0.
datalab-toopen-source9B paramsVision-Language OCR ModelApache 2.02 current SOTA
#1 on olmOCR-Bench (83.1). Best on old scans math, long tiny text, base accuracy.
§ 01 · Benchmarks
Every benchmark Chandra v0.1.0 has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | olmOCR-Bench | Computer Vision · Document Parsing | long-tiny-text | 92.3% | #1 | — | source ↗ |
| 02 | olmOCR-Bench | Computer Vision · Document Parsing | base | 99.9% | #1 | — | source ↗ |
| 03 | olmOCR-Bench | Computer Vision · Document Parsing | old-scans | 50.4% | #2 | — | source ↗ |
| 04 | olmOCR-Bench | Computer Vision · Document Parsing | headers-footers | 90.8% | #3 | — | source ↗ |
| 05 | olmOCR-Bench | Computer Vision · Document Parsing | pass-rate | 83.1% | #3 | — | source ↗ |
| 06 | olmOCR-Bench | Computer Vision · Document Parsing | old-scans-math | 80.3% | #3 | — | source ↗ |
| 07 | olmOCR-Bench | Computer Vision · Document Parsing | tables | 88.0% | #3 | — | source ↗ |
| 08 | olmOCR-Bench | Computer Vision · Document Parsing | multi-column | 81.2% | #4 | — | source ↗ |
| 09 | olmOCR-Bench | Computer Vision · Document Parsing | arxiv | 82.2% | #4 | — | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 02 · Strengths by area
Where Chandra v0.1.0 actually performs.
§ 05 · Sources & freshness
Where these numbers come from.
github-readme
8
results
alphaxiv-leaderboard
1
result
0 of 9 rows marked verified.