olmOCR-BenchParseBenchfr-bench-pdf2mdGlobal average

fr-bench-pdf2md

Overall score versus model parameters. Smaller and higher is better.

fr-bench-pdf2md score versus model sizeLower parameter count and higher overall score are better. Dashed line: Pareto frontier.45505560657075800.712481632Model parameters (B) · Smaller is betterOverall score · Higher is betterMistral OCR 4.1 — 54.06 points — Size undisclosed — Reported resultsMistral OCR 4.1 · 54.1LightOnOCR-3-4B — 74.1 points — 4.0B — Evaluated outputLightOnOCR-3 4BLightOnOCR-3-0.8B — 70.5 points — 0.8B — Evaluated outputLightOnOCR-3 0.8BLightOnOCR-3-1B — 69.6 points — 1.0B — Evaluated outputLightOnOCR-3-1Bchandra-ocr-2 — 69.0 points — 4.0B — Evaluated outputchandra-ocr-2Infinity-Parser2-Pro — 63.2 points — 35.1B — Reported resultsInfinity-Parser2-Prodots.mocr — 59.3 points — 3.0B — Evaluated outputdots.mocrsurya-ocr-2 — 54.1 points — 0.7B — Evaluated outputsurya-ocr-2jina-ocr-v1 — 53.2 points — 3.4B — Evaluated outputjina-ocr-v1
External modelsLightOnOCR-2LightOnOCR-3Pareto frontier
Hover or focus a point for its exact checkpoint and scores.
Method and source

Smaller size and higher score are better. The frontier includes points that no other comparable point matches or improves on in both dimensions, with a strict improvement in at least one. Comparisons use source precision; displayed scores are rounded to one decimal. Parameter counts are those in the tables, including nominal backbone sizes where specified.

Overall uses five categories for ParseBench and includes headers/footers for olmOCR. All points match the selected table rows; LightOnOCR-3 uses the candidate-1 evaluations for 0.8B, 1B, and 4B. Postprocessing applies only to olmOCR; ParseBench and fr-bench-pdf2md outputs are evaluated as submitted. The same candidate checkpoints are used across all three benchmarks. Missing scores or sizes are omitted. Models with undisclosed parameter counts (such as MistralOCR4.1) cannot be placed on a size axis. * LightOnOCR-2 uses its published olmOCR score excluding headers/footers; it is shown but excluded from the frontier. Mistral OCR 4 is a horizontal score reference because its size is undisclosed; it is excluded from the size frontier. Its olmOCR score is the user-provided reported 85.20; ParseBench and fr-bench-pdf2md retain their existing selected scores.

Leaderboard source revision: 3576c5240602588f596f8790a3a7161a524e59ad. Values are pinned to this snapshot.

Show scores and output details
Plotted scores
ModelSize (B)OverallOutput / inputsFrontier
LightOnOCR-3-4B4.074.1Evaluated output✓
LightOnOCR-3-0.8B0.870.5Evaluated output✓
LightOnOCR-3-1B1.069.6Evaluated output—
chandra-ocr-24.069.0Evaluated output—
Infinity-Parser2-Pro35.163.2Reported results—
dots.mocr3.059.3Evaluated output—
surya-ocr-20.754.1Evaluated output✓
Mistral OCR 4.1Undisclosed54.1Reported resultsSize undisclosed
jina-ocr-v13.453.2Evaluated output—