{"id":1184834,"date":"2026-08-31T09:00:00","date_gmt":"2026-08-31T16:00:00","guid":{"rendered":""},"modified":"2026-08-31T06:43:23","modified_gmt":"2026-08-31T13:43:23","slug":"gigapath-flash-and-gigatime-flash-toward-population-scale-discovery-with-efficient-pathology-foundation-models","status":"publish","type":"post","link":"https:\/\/www.microsoft.com\/en-us\/research\/blog\/gigapath-flash-and-gigatime-flash-toward-population-scale-discovery-with-efficient-pathology-foundation-models\/","title":{"rendered":"GigaPath-Flash and GigaTIME-Flash: Toward population-scale discovery with efficient pathology foundation models"},"content":{"rendered":"\n<figure class=\"wp-block-image size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"1400\" height=\"788\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW.jpg\" alt=\"Overview of the GigaPath\/GigaTIME model family. GigaPath-Flash provides efficient tile and slide encoders at 22M parameters. GigaTIME-Flash predicts spatial proteomics from H&E, replacing the CNN backbone with the distilled ViT-S encoder. \" class=\"wp-image-1184911\" srcset=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW.jpg 1400w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-300x169.jpg 300w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-1024x576.jpg 1024w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-768x432.jpg 768w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-1066x600.jpg 1066w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-655x368.jpg 655w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-240x135.jpg 240w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-640x360.jpg 640w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-960x540.jpg 960w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-1280x720.jpg 1280w\" sizes=\"auto, (max-width: 1400px) 100vw, 1400px\" \/><\/figure>\n\n\n\n<div style=\"padding-bottom:0;padding-top:0\" class=\"wp-block-msr-immersive-section alignfull row\">\n\t\n\t<div class=\"container\">\n\t\t<div class=\"wp-block-msr-immersive-section__inner wp-block-msr-immersive-section__inner--narrow\">\n\t\t\t<div class=\"wp-block-columns mb-10 pb-1 pr-1 is-layout-flex wp-container-core-columns-is-layout-8f761849 wp-block-columns-is-layout-flex\" style=\"box-shadow:var(--wp--preset--shadow--outlined)\">\n<div class=\"wp-block-column is-layout-flow wp-block-column-is-layout-flow\">\n<h2 id=\"at-a-glance\" class=\"wp-block-heading h3\">At a glance<\/h2>\n\n\n\n<ul class=\"wp-block-list\">\n<li>The Flash family extends GigaPath and GigaTIME with dramatically improved efficiency, making large-scale pathology research more accessible and practical.<\/li>\n\n\n\n<li>A distilled pathology foundation model backbone reduces computational requirements without sacrificing performance, enabling repeated analyses across larger patient cohorts.<\/li>\n\n\n\n<li>These open models support population-scale discovery, helping researchers investigate disease biology, biomarkers, and clinical outcomes across diverse cancer datasets.<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\"><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/aka.ms\/gigapath\" target=\"_blank\" rel=\"noopener noreferrer\"><em>GigaPath<\/em><span class=\"sr-only\"> (opens in new tab)<\/span><\/a><em> and <\/em><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/aka.ms\/gigatime\" target=\"_blank\" rel=\"noopener noreferrer\"><em>GigaTIME<\/em><span class=\"sr-only\"> (opens in new tab)<\/span><\/a><em> demonstrated how foundation models can support whole-slide analysis and tumor microenvironment modeling from routinely collected pathology data. GigaPath-Flash and GigaTIME-Flash make these capabilities substantially more efficient, enabling researchers to analyze larger cohorts, run more experiments, and move toward population-scale discovery.<\/em>\u00a0<em>GigaPath-Flash and GigaTIME-Flash are research models. They are not intended or validated for clinical use, including diagnosis, prognosis, treatment selection, or other patient-care decisions. Performance may vary across datasets, scanners, institutions, populations, and use cases.<\/em><\/p>\n<\/div>\n<\/div>\t\t<\/div>\n\t<\/div>\n\n\t<\/div>\n\n\n\n<h2 id=\"the-scale-opportunity-in-computational-pathology\" class=\"wp-block-heading\">The scale opportunity in computational pathology<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Histopathology is among the richest and most widely available sources of information in cancer research. Every tissue biopsy produces a whole-slide image that captures cellular morphology at subcellular resolution \u2014 and hospitals generate millions of these slides each year. This data contains information relevant to diagnosis, prognosis, treatment selection, and the biology of the tumor microenvironment.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Foundation models have begun to unlock this information at scale. But whole-slide images are large \u2014 often exceeding a gigapixel \u2014 and applying a foundation model to even a single slide requires processing thousands of image tiles. When a research question involves tens of thousands of patients, the computational cost grows quickly. And population-scale discovery is not a single model run: it requires repeated cycles of feature extraction, statistical analysis, hypothesis testing, and validation across patient subgroups, biomarkers, and clinical endpoints.\u00a0<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Computational cost limits the number of patients, datasets, tasks, and hypotheses that researchers can study. To realize the full potential of pathology foundation models, we need models that can be applied repeatedly and affordably across large patient populations.<\/p>\n\n\n\n<h2 id=\"from-gigapath-and-gigatime-to-the-flash-family\" class=\"wp-block-heading\">From GigaPath and GigaTIME to the -Flash family<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\"><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/aka.ms\/gigapath-paper\" type=\"link\" id=\"https:\/\/aka.ms\/gigapath-paper\" target=\"_blank\" rel=\"noopener noreferrer\">GigaPath (Nature, 2024)<span class=\"sr-only\"> (opens in new tab)<\/span><\/a> is a whole-slide foundation model pretrained on large-scale real-world histopathology data from Providence. Unlike models that operate only at the tile level, GigaPath learns contextualized representations of entire slides, capturing both local cellular patterns and global tissue architecture.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/aka.ms\/gigatime-paper\" type=\"link\" id=\"https:\/\/aka.ms\/gigatime-paper\" target=\"_blank\" rel=\"noopener noreferrer\">GigaTIME (Cell, 2026)<span class=\"sr-only\"> (opens in new tab)<\/span><\/a> extends this line of work to tumor microenvironments. Trained on 40 million cells with paired H&E and multiplex immunofluorescence (mIF) data, GigaTIME translates routine H&E images into virtual spatial proteomics maps across 21 protein channels. Applied to over 14,000 cancer patients, it generated a virtual population that uncovered more than 1,200 statistically significant associations between immune cell states and clinical biomarkers.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">GigaPath and GigaTIME addressed the scale of pathology data and biological discovery. And now, the Flash family of models addresses scale of experimentation.<\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"980\" height=\"417\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig1_overview_GPF_GTF.png\" alt=\"Figure 1: Overview of the GigaPath\/GigaTIME model family. GigaPath-Flash provides efficient tile and slide encoders at 22M parameters. GigaTIME-Flash predicts spatial proteomics from H&E, replacing the CNN backbone with the distilled ViT-S encoder. \" class=\"wp-image-1184838\" srcset=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig1_overview_GPF_GTF.png 980w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig1_overview_GPF_GTF-300x128.png 300w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig1_overview_GPF_GTF-768x327.png 768w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig1_overview_GPF_GTF-240x102.png 240w\" sizes=\"auto, (max-width: 980px) 100vw, 980px\" \/><figcaption class=\"wp-element-caption\">Figure 1: Overview of the GigaPath\/GigaTIME model family. GigaPath-Flash provides efficient tile and slide encoders at 22M parameters. GigaTIME-Flash predicts spatial proteomics from H&E, replacing the CNN backbone with the distilled ViT-S encoder.<\/figcaption><\/figure>\n\n\n\n<h2 id=\"introducing-gigapath-flash-and-gigatime-flash\" class=\"wp-block-heading\">Introducing GigaPath-Flash and GigaTIME-Flash<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">The Flash family shares a core design goal: preserving useful pathology representations while substantially reducing the computational resources required to generate and use them. Both models are built on a common efficient backbone \u2014 a compact ViT-S tile encoder distilled from the original billion-parameter GigaPath encoder \u2014 and both are released under the Apache 2.0 license.\u00a0<\/p>\n\n\n\n<h3 id=\"gigapath-flash\" class=\"wp-block-heading\">GigaPath-Flash<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">GigaPath-Flash is an efficient foundation model for whole-slide representation learning. It combines a 22M-parameter ViT-S tile encoder with a 21M-parameter LongNet slide encoder. The tile encoder is distilled from the original GigaPath ViT-g teacher, transferring the representational capacity of a billion-parameter model into a backbone that is an order of magnitude smaller. The slide encoder contextualizes all tile embeddings via dilated attention, scaling linearly with the number of tiles.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">On slide-level classification benchmarks (PANDA prostate grading and EBRAINS brain tumor subtyping), GigaPath-Flash achieves the lowest inference cost among whole-slide pretrained models while retaining competitive performance \u2014 scoring within 3% of the original GigaPath at roughly 50 times less compute.<\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"847\" height=\"536\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig2_gigapath_slide_benchmark_efficiency_GPF_GTF.png\" alt=\"Figure 2: Efficiency\u2013performance trade-off on whole-slide benchmarks. GigaPath-Flash (red, top-left) achieves competitive performance at substantially lower computational cost than other whole-slide pretrained models.\" class=\"wp-image-1184839\" srcset=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig2_gigapath_slide_benchmark_efficiency_GPF_GTF.png 847w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig2_gigapath_slide_benchmark_efficiency_GPF_GTF-300x190.png 300w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig2_gigapath_slide_benchmark_efficiency_GPF_GTF-768x486.png 768w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig2_gigapath_slide_benchmark_efficiency_GPF_GTF-240x152.png 240w\" sizes=\"auto, (max-width: 847px) 100vw, 847px\" \/><figcaption class=\"wp-element-caption\">Figure 2: Efficiency\u2013performance trade-off on whole-slide benchmarks. GigaPath-Flash (red, top-left) achieves competitive performance at substantially lower computational cost than other whole-slide pretrained models.<\/figcaption><\/figure>\n\n\n\n<h3 id=\"gigatime-flash\" class=\"wp-block-heading\">GigaTIME-Flash<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">GigaTIME-Flash replaces the CNN backbone of the original GigaTIME with the GigaPath-Flash ViT-S encoder, paired with a lightweight convolutional decoder for H&E-to-mIF translation. The model is fine-tuned using LoRA adapters, keeping the pretrained encoder weights largely frozen.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">On both in-distribution and out-of-distribution cohorts spanning brain, breast, colon, and lung cancers, GigaTIME-Flash matches or improves upon the original GigaTIME in spatial protein prediction quality. The gains are particularly notable on out-of-distribution data, suggesting that the foundation model backbone improves generalization to previously unseen tissue types.<\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"1203\" height=\"612\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig3_gigatime_mean_pearson_bars_GPF_GTF.png\" alt=\"Figure 3: Mean windowed Pearson correlation for GigaTIME and GigaTIME-Flash on the GigaTIME test set and four out-of-distribution Prov-TMA cohorts. GigaTIME-Flash matches or improves upon the original across all cohorts.\" class=\"wp-image-1184840\" srcset=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig3_gigatime_mean_pearson_bars_GPF_GTF.png 1203w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig3_gigatime_mean_pearson_bars_GPF_GTF-300x153.png 300w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig3_gigatime_mean_pearson_bars_GPF_GTF-1024x521.png 1024w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig3_gigatime_mean_pearson_bars_GPF_GTF-768x391.png 768w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig3_gigatime_mean_pearson_bars_GPF_GTF-240x122.png 240w\" sizes=\"auto, (max-width: 1203px) 100vw, 1203px\" \/><figcaption class=\"wp-element-caption\">Figure 3: Mean windowed Pearson correlation for GigaTIME and GigaTIME-Flash on the GigaTIME test set and four out-of-distribution Prov-TMA cohorts. GigaTIME-Flash matches or improves upon the original across all cohorts.<\/figcaption><\/figure>\n\n\n\n<h2 id=\"efficiency-without-giving-up-the-foundation\" class=\"wp-block-heading\">Efficiency without giving up the foundation<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">The efficiency gains of the Flash models are substantial:<\/p>\n\n\n\n<figure class=\"wp-block-table\"><table><thead><tr><th>Model<\/th><th>Type<\/th><th>Efficiency gain<\/th><\/tr><\/thead><tbody><tr><td>GigaPath-Flash<\/td><td>Whole-slide Foundation Model<\/td><td>~50\u00d7 less compute, 97% of predictive performance compared to GigaPath<\/td><\/tr><tr><td>GigaTIME-Flash<\/td><td>Spatial Proteomics<\/td><td>~6\u00d7 faster, ~8\u00d7 less memory, better predictive performance compared to GigaTIME<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">For a single slide, these differences reduce runtime and hardware requirements. Across tens of thousands of slides, they can determine whether an experiment is practical at all. To illustrate, we estimate the wall-clock time for generating virtual mIF across cohorts of different sizes on a single A100 GPU, assuming approximately 10,000 tiles per slide:<\/p>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><th>Model<\/th><th>1,000 slides<\/th><th>100K slides<\/th><th>1M slides<\/th><\/tr><\/thead><tbody><tr><td><strong>GigaTIME-Flash<\/strong><\/td><td>~2 GPU-hours<\/td><td>~7 GPU-days<\/td><td>~70 GPU-days<\/td><\/tr><tr><td>GigaTIME<\/td><td>~7 GPU-hours<\/td><td>~30 GPU-days<\/td><td>~300 GPU-days<\/td><\/tr><\/tbody><\/table><figcaption class=\"wp-element-caption\">Estimates assume ~10,000 tiles per slide, batch size 128, single NVIDIA A100 GPU. Actual runtime depends on slide size, tiling resolution, and hardware.<\/figcaption><\/figure>\n\n\n\n<figure class=\"wp-block-image aligncenter size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"842\" height=\"375\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig4_gigatime_efficiency_plot_GPF_GTF.png\" alt=\"Figure 4: GigaTIME efficiency scaling. Left: throughput (tiles\/sec) vs. batch size. Right: peak GPU memory (GB) vs. batch size. GigaTIME-Flash scales to over 1,600 tiles\/sec while using a fraction of the memory.\" class=\"wp-image-1184841\" srcset=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig4_gigatime_efficiency_plot_GPF_GTF.png 842w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig4_gigatime_efficiency_plot_GPF_GTF-300x134.png 300w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig4_gigatime_efficiency_plot_GPF_GTF-768x342.png 768w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/fig4_gigatime_efficiency_plot_GPF_GTF-240x107.png 240w\" sizes=\"auto, (max-width: 842px) 100vw, 842px\" \/><figcaption class=\"wp-element-caption\">Figure 4: GigaTIME efficiency scaling. Left: throughput (tiles\/sec) vs. batch size. Right: peak GPU memory (GB) vs. batch size. GigaTIME-Flash scales to over 1,600 tiles\/sec while using a fraction of the memory.<\/figcaption><\/figure>\n\n\n\n<h2 id=\"an-open-model-release\" class=\"wp-block-heading\">An open model release<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Both GigaPath-Flash and GigaTIME-Flash are released as open-weight models under the Apache 2.0 license. Model weights and code are available on HuggingFace:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/aka.ms\/GigaPath-Flash\" type=\"link\" id=\"aka.ms\/GigaPath-Flash\" target=\"_blank\" rel=\"noopener noreferrer\">GigaPath-Flash<span class=\"sr-only\"> (opens in new tab)<\/span><\/a><\/li>\n\n\n\n<li><a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" href=\"https:\/\/aka.ms\/GigaTIME-Flash\" type=\"link\" id=\"aka.ms\/GigaTIME-Flash\" target=\"_blank\" rel=\"noopener noreferrer\">GigaTIME-Flash<span class=\"sr-only\"> (opens in new tab)<\/span><\/a><\/li>\n\n\n\n<li><a href=\"https:\/\/www.microsoft.com\/en-us\/research\/publication\/gigapath-flash-and-gigatime-flash-efficient-pathology-foundation-models-for-whole-slide-and-tumor-microenvironment-analysis\/\" type=\"msr-research-item\" id=\"1180651\">GigaPath-Flash and GigaTIME-Flash: Efficient Pathology Foundation Models for Whole-Slide and Tumor Microenvironment Analysis<\/a><\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">This is an early research release. Our current evaluations cover a limited set of benchmarks and cohorts, and broader validation across tasks, scanners, and patient populations is still needed. We expect the most valuable applications of these models to include scientific questions, cohorts, and use cases beyond those in our initial experiments. We welcome community evaluation, and we are equally interested in reports of where these models work well and where they fall short.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Downstream clinical applications will require additional multi-institutional and prospective validation.<\/p>\n\n\n\n\t<div class=\"border-bottom border-top border-gray-300 mt-5 mb-5 msr-promo text-center text-md-left alignwide\" data-bi-aN=\"promo\" data-bi-id=\"1141385\">\n\t\t\n\n\t\n\t<div class=\"row pt-3 pb-4 align-items-center\">\n\t\t\t\t\t\t<div class=\"msr-promo__media col-12 col-md-5\">\n\t\t\t\t<a class=\"bg-gray-300 display-block\" href=\"https:\/\/ai.azure.com\/labs\" aria-label=\"Azure AI Foundry Labs\" data-bi-cn=\"Azure AI Foundry Labs\" target=\"_blank\">\n\t\t\t\t\t<img decoding=\"async\" class=\"w-100 display-block\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2025\/06\/Azure-AI-Foundry_1600x900.jpg\" alt=\"decorative image and the text \" \/>\n\t\t\t\t<\/a>\n\t\t\t<\/div>\n\t\t\t\n\t\t\t<div class=\"msr-promo__content p-3 px-5 col-12 col-md\">\n\n\t\t\t\t\t\t\t\t\t<h2 class=\"h4\">Azure AI Foundry Labs<\/h2>\n\t\t\t\t\n\t\t\t\t\t\t\t\t<p id=\"azure-ai-foundry-labs\" class=\"large\">Get a glimpse of potential future directions for AI, with these experimental technologies from Microsoft Research.<\/p>\n\t\t\t\t\n\t\t\t\t\t\t\t\t<div class=\"wp-block-buttons justify-content-center justify-content-md-start\">\n\t\t\t\t\t<div class=\"wp-block-button\">\n\t\t\t\t\t\t<a href=\"https:\/\/ai.azure.com\/labs\" aria-describedby=\"azure-ai-foundry-labs\" class=\"btn btn-brand glyph-append glyph-append-chevron-right\" data-bi-cn=\"Azure AI Foundry Labs\" target=\"_blank\">\n\t\t\t\t\t\t\tAzure AI Foundry\t\t\t\t\t\t<\/a>\n\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t\t\t<\/div><!--\/.msr-promo__content-->\n\t<\/div><!--\/.msr-promo__inner-wrap-->\n\t<\/div><!--\/.msr-promo-->\n\t\n\n\n<h2 id=\"efficiency-as-an-enabler-of-discovery\" class=\"wp-block-heading\">Efficiency as an enabler of discovery<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">GigaPath and GigaTIME demonstrated what pathology foundation models can learn from whole slides and tumor tissue. GigaPath-Flash and GigaTIME-Flash are a step toward making those capabilities usable across larger populations, more experiments, and a broader research community.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">By making pathology foundation models more efficient, we hope to expand the scale of the scientific questions researchers can ask.<\/p>\n\n\n\n<h2 id=\"acknowledgements\" class=\"wp-block-heading\">Acknowledgements<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">GigaPath-Flash and GigaTIME-Flash are joint work across Microsoft Research, the University of Washington, and Providence. For technical details, see the <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/publication\/gigapath-flash-and-gigatime-flash-efficient-pathology-foundation-models-for-whole-slide-and-tumor-microenvironment-analysis\/\" type=\"msr-research-item\" id=\"1180651\">paper<\/a>.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><em><strong>Paper co-authors<\/strong>: Naoto Usuyama, Jeya Maria Jose Valanarasu, Sicong Yao, Hanwen Xu, Jaspreet Bagga, Guanghui Qin, Robert E. Kramer, Cliff Wong, Soohee Lee, Hao Qiu, Theodore Zhengde Zhao, Racheli Ben Shimol, Angela Crabtree, Kevin Matlock, Eduardo Alejandro Lozano Garcia, Naiteek Sangani, Alberto Santamaria-Pang, Maximilian Rokuss, Yashna Hasija, Naisargi Manishkumar Patel, Jason Entenmann, Alexandra Q. Bartlett, Bill J. Wright, Bernard A. Fox, Brian Piening, Sheng Zhang, Sheng Wang, Tristan Naumann, Carlo Bifulco, Hoifung Poon<\/em><\/p>\n","protected":false},"excerpt":{"rendered":"<p>What if pathology foundation models could do more with less? GigaPath-Flash and GigaTIME-Flash cut computational demands while maintaining strong performance, opening the door to larger studies and broader exploration.<\/p>\n","protected":false},"author":43868,"featured_media":1184911,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr-author-ordering":[{"type":"user_nicename","value":"Naoto Usuyama","user_id":"38670"},{"type":"user_nicename","value":"Jeya Maria Jose Valanarasu","user_id":"43491"},{"type":"user_nicename","value":"Tristan Naumann","user_id":"37929"}],"msr_hide_image_in_river":0,"footnotes":""},"categories":[1],"tags":[],"research-area":[13553],"msr-region":[],"msr-event-type":[],"msr-locale":[268875],"msr-post-option":[243984],"msr-impact-theme":[],"msr-promo-type":[],"msr-podcast-series":[],"class_list":["post-1184834","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-research-blog","msr-research-area-medical-health-genomics","msr-locale-en_us","msr-post-option-blog-homepage-featured"],"msr_event_details":{"start":"","end":"","location":""},"podcast_url":"","podcast_episode":"","msr_research_lab":[849856],"msr_impact_theme":[],"related-publications":[],"related-downloads":[],"related-videos":[],"related-academic-programs":[],"related-groups":[],"related-projects":[],"related-events":[],"related-researchers":[{"type":"user_nicename","value":"Naoto Usuyama","user_id":38670,"display_name":"Naoto Usuyama","author_link":"<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/naotous\/\" aria-label=\"Visit the profile page for Naoto Usuyama\">Naoto Usuyama<\/a>","is_active":false,"last_first":"Usuyama, Naoto","people_section":0,"alias":"naotous"},{"type":"user_nicename","value":"Jeya Maria Jose Valanarasu","user_id":43491,"display_name":"Jeya Maria Jose Valanarasu","author_link":"<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/jevalanarasu\/\" aria-label=\"Visit the profile page for Jeya Maria Jose Valanarasu\">Jeya Maria Jose Valanarasu<\/a>","is_active":false,"last_first":"Valanarasu, Jeya Maria Jose","people_section":0,"alias":"jevalanarasu"},{"type":"user_nicename","value":"Tristan Naumann","user_id":37929,"display_name":"Tristan Naumann","author_link":"<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/tristan\/\" aria-label=\"Visit the profile page for Tristan Naumann\">Tristan Naumann<\/a>","is_active":false,"last_first":"Naumann, Tristan","people_section":0,"alias":"tristan"}],"msr_type":"Post","featured_image_thumbnail":"<img width=\"960\" height=\"540\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-960x540.jpg\" class=\"img-object-cover\" alt=\"Overview of the GigaPath\/GigaTIME model family. GigaPath-Flash provides efficient tile and slide encoders at 22M parameters. GigaTIME-Flash predicts spatial proteomics from H&amp;E, replacing the CNN backbone with the distilled ViT-S encoder.\" decoding=\"async\" loading=\"lazy\" srcset=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-960x540.jpg 960w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-300x169.jpg 300w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-1024x576.jpg 1024w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-768x432.jpg 768w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-1066x600.jpg 1066w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-655x368.jpg 655w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-240x135.jpg 240w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-640x360.jpg 640w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW-1280x720.jpg 1280w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/08\/GigaPathFlash-BlogHeroFeature-1400x788_NEW.jpg 1400w\" sizes=\"auto, (max-width: 960px) 100vw, 960px\" \/>","byline":"<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/naotous\/\" title=\"Go to researcher profile for Naoto Usuyama\" aria-label=\"Go to researcher profile for Naoto Usuyama\" data-bi-type=\"byline author\" data-bi-cN=\"Naoto Usuyama\">Naoto Usuyama<\/a>, <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/jevalanarasu\/\" title=\"Go to researcher profile for Jeya Maria Jose Valanarasu\" aria-label=\"Go to researcher profile for Jeya Maria Jose Valanarasu\" data-bi-type=\"byline author\" data-bi-cN=\"Jeya Maria Jose Valanarasu\">Jeya Maria Jose Valanarasu<\/a>, and <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/people\/tristan\/\" title=\"Go to researcher profile for Tristan Naumann\" aria-label=\"Go to researcher profile for Tristan Naumann\" data-bi-type=\"byline author\" data-bi-cN=\"Tristan Naumann\">Tristan Naumann<\/a>","formattedDate":"August 31, 2026","formattedExcerpt":"What if pathology foundation models could do more with less? GigaPath-Flash and GigaTIME-Flash cut computational demands while maintaining strong performance, opening the door to larger studies and broader exploration.","locale":{"slug":"en_us","name":"English","native":"","english":"English"},"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/posts\/1184834","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/users\/43868"}],"replies":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/comments?post=1184834"}],"version-history":[{"count":14,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/posts\/1184834\/revisions"}],"predecessor-version":[{"id":1184933,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/posts\/1184834\/revisions\/1184933"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media\/1184911"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=1184834"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/categories?post=1184834"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/tags?post=1184834"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=1184834"},{"taxonomy":"msr-region","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-region?post=1184834"},{"taxonomy":"msr-event-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event-type?post=1184834"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=1184834"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=1184834"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=1184834"},{"taxonomy":"msr-promo-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-promo-type?post=1184834"},{"taxonomy":"msr-podcast-series","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-podcast-series?post=1184834"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}