PBSBench - Cell Fill-in-the-Blank: leaderboard
PBSBench (2026) tests vision-language models on peripheral blood smear hematopathology, with questions about single white blood cell crops. Share of fill-in-the-blank answers that contain the reference keyword, in percent, on in-domain cells; higher is better.
Metric: Partial match rate (answer contains the ground-truth keyword after case and whitespace normalization) (%, times 100) on PBSBench cell-level in-domain fill-in-the-blank questions about single white-blood-cell crops from peripheral blood smears (916 questions); questions generated from GPT-4o morphology captions of hematopathologist-labeled cells; greedy decoding, model chat templates; higher is better. Source: arxiv.org. Saturation forecast: Around February 2027. 16 models tracked. Updated 7 Oct 2026.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.5 | 54 | #261 |
| 2 | GPT-4o | 48 | #751 |
| 3 | Gemini 2.5 Pro | 41 | #276 |
| 4 | InternVL3.5-8B | 38 | #1604 |
| 5 | PathGen-LLaVA | 31 | #1406 |
| 6 | Qwen 2 VL 7B Instruct | 27 | #2573 |
| 7 | BLIP2-OPT-6.7B | 21 | |
| 8 | Qwen 2.5 VL 7B Instruct | 17 | #1902 |
| 9 | Idefics3-8B-Llama3 | 17 | #5542 |
| 10 | Quilt-LLaVA | 16 | #4973 |
| 11 | DeepSeek-VL-7B | 15 | #7399 |
| 12 | llava-v1.6-mistral-7B | 10 | #3999 |
| 13 | LLaVA-Med | 10 | |
| 14 | LLaVA-1.6-Vicuna-7B | 10 | |
| 15 | Med-Flamingo | 6 |
No result here: #1 Claude Fable 5.2, #3 Gemini 4 Neon, #4 GPT-6.2 Astra.
Interactive version: theaggregate.ai/benchmark?slug=pbsbench-cell-fill-in-the-blank · How It Works · Data refreshed daily, snapshot 2026-11-21.