ReceiptBench - Normalization: leaderboard

Formatting and normalization fields (dates and totals in a fixed format) of ReceiptBench (2026), structured extraction of 19 fields from photos and scans of real receipts and invoices (purchase, dining, travel, hotel, taxi and more), 2,000 held-out test images, JSON output. Field-level F1 times 100 (0-100), exact, numeric or list matching with a small LLM judge for near misses; higher is better.

Metric: Formatting and normalization fields (dates as YYYY-MM-DD, totals as decimals), field-level F1 (times 100, so 0-100) on the 2,000-image held-out test set of ReceiptBench (real travel and expense receipts and invoices, 19 annotated fields), one-shot schema prompt with JSON output; fields are scored by exact, numeric or structured-list matching with a Qwen3-4B semantic judge when string similarity falls below the match threshold; higher is better. Source: arxiv.org. Saturation forecast: Estimated already saturated. 14 models tracked. Updated 7 Oct 2026.

Top models

#ModelScoreOverall rank
1Gemini 3 Pro90.86#146
2Qwen 3 VL Plus90#514
3InternVL3.5-241B87.91#632
4GPT-587.43#185
5Qwen 3 VL 8B85.77#927
6InternVL3-78B84.44#765
7Qwen 3 VL 32B83.1#624
8PaddleOCR-VL81.25
9olmOCR-7B80.3
10InternVL3-8B76.61#1795
11Qwen 3 VL 4B75.03#1163
12DianJin-OCR-R171.26
13InternVL3-2B58.21#3318
14DeepSeek-OCR-small50.81

No result here: #2 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.

Interactive version: theaggregate.ai/benchmark?slug=receiptbench-normalization · How It Works · Data refreshed daily, snapshot 2026-11-22.