Compugen searches for immuno-oncology drug targets using massive single-cell and tumor multi-omics datasets. Its Unigen platform integrates multi-omics, spatial and single-cell data to identify new drug targets and resistance pathways. It began collaborating with Nebius through an Academy prototype. Academy graduates worked with Compugen specialists on a TLS classification project using scGPT, scGPT-Spatial and Nicheformer across 14 cohorts and hundreds of thousands of cells. Nebius supplied the project team with a special GPU Cloud grant.
Compugen then used Nebius AI Cloud and NVIDIA GPUs to train a production-grade research model that predicts spatial immune features, including tertiary lymphoid structures, from tumor data. Nebius describes training large models, including single-cell RNA transformers, with models up to 3B parameters running in BF16/FP32. Its life-sciences page also reports fine-tuning on proprietary cancer datasets and training times of 20 seconds to 20 minutes per epoch for 100K-cell RNA-seq models.
Compugen’s research includes a variant designed to infer these features from expression data without spatial coordinates. Nebius reports a best TLS prediction result of 0.94 ROC-AUC without spatial coordinates. Compugen described integrating the models into its Unigen discovery pipelines as a next step. Nebius also describes its infrastructure as helping Compugen integrate AI into that pipeline.
Nebius describes helping Compugen train cancer research models, including single-cell RNA transformers with models up to 3B parameters running in BF16/FP32. It reports a best TLS prediction result of 0.94 ROC-AUC without spatial coordinates and says Nebius helped integrate AI into Compugen’s Unigen discovery pipeline.
Cloud for Scientific AI and Healthcare13 Oct 2026 · nebius.com · Web pageNebius describes helping Compugen train cancer research models, including single-cell RNA transformers with models up to 3B parameters running in BF16/FP32. It reports a best TLS prediction result of 0.94 ROC-AUC without spatial coordinates and says Nebius helped integrate AI into Compugen’s Unigen discovery pipeline.
Nebius Customer story
27 Sept 2026 · Nebius · Web page
Nebius describes Compugen’s collaboration expanding from an educational partnership at Nebius Academy into full-scale training on Nebius AI Cloud. The work produced a research model for tertiary lymphoid structure prediction using scGPT, scGPT-Spatial and Nicheformer.
Nebius Customer story27 Sept 2026 · Nebius · Web pageNebius describes Compugen’s collaboration expanding from an educational partnership at Nebius Academy into full-scale training on Nebius AI Cloud. The work produced a research model for tertiary lymphoid structure prediction using scGPT, scGPT-Spatial and Nicheformer.
#aiinhealthcare #cancerresearch #nebiusacademy | Nebius Academy Israel
13 Oct 2026 · www.linkedin.com · Web page
Nebius Academy Israel names Compugen as its industry partner for a graduate TLS classification project. Graduates worked with Compugen specialists using scGPT, scGPT-Spatial and Nicheformer across 14 cohorts and hundreds of thousands of cells. Nebius supplied the Academy project team with a special GPU Cloud grant.
#aiinhealthcare #cancerresearch #nebiusacademy | Nebius Academy Israel13 Oct 2026 · www.linkedin.com · Web pageNebius Academy Israel names Compugen as its industry partner for a graduate TLS classification project. Graduates worked with Compugen specialists using scGPT, scGPT-Spatial and Nicheformer across 14 cohorts and hundreds of thousands of cells. Nebius supplied the Academy project team with a special GPU Cloud grant.
Compugen is a clinical-stage biotechnology company developing cancer immunotherapies. Its Unigen platform uses AI, machine learning and biological data to identify drug targets and guide antibody development. Its pipeline includes COM701, an anti-PVRIG antibody being studied as maintenance therapy for relapsed ovarian cancer; rilvegostomig, a PD-1/TIGIT bispecific antibody developed by AstraZeneca using a component derived from Compugen’s COM902; and GS-0321, an antibody licensed to Gilead and undergoing a Phase 1 trial. Its programs target cancer patients and support pharmaceutical partners’ drug development.
At its October 5 platform presentation, Compugen introduced UniGenX, an expansion of its AI discovery engine focused on tumor-associated antigens. It combines multiple biological data types to prioritize targets for cancer therapies, with candidates undergoing further experimental validation.
Compugen reports second-quarter revenue of $2.6 million and $7 million net loss
Compugen reported second-quarter revenue of approximately $2.6 million, up from $1.3 million a year earlier, and a net loss of $7 million. Cash and investments totaled $125.3 million at June 30, with operations expected to be funded into 2029. Interim progression-free survival data from its COM701 MAIA-ovarian trial remained expected by the first quarter of 2027.
Compugen searches for immuno-oncology drug targets using massive single-cell and tumor multi-omics datasets. Its Unigen platform integrates multi-omics, spatial and single-cell data to identify new drug targets and resistance pathways. It began collaborating with Nebius through an Academy prototype. Academy graduates worked with Compugen specialists on a TLS classification project using scGPT, scGPT-Spatial and Nicheformer across 14 cohorts and hundreds of thousands of cells. Nebius supplied the project team with a special GPU Cloud grant.
Compugen then used Nebius AI Cloud and NVIDIA GPUs to train a production-grade research model that predicts spatial immune features, including tertiary lymphoid structures, from tumor data. Nebius describes training large models, including single-cell RNA transformers, with models up to 3B parameters running in BF16/FP32. Its life-sciences page also reports fine-tuning on proprietary cancer datasets and training times of 20 seconds to 20 minutes per epoch for 100K-cell RNA-seq models.
Compugen’s research includes a variant designed to infer these features from expression data without spatial coordinates. Nebius reports a best TLS prediction result of 0.94 ROC-AUC without spatial coordinates. Compugen described integrating the models into its Unigen discovery pipelines as a next step. Nebius also describes its infrastructure as helping Compugen integrate AI into that pipeline.
Nebius describes helping Compugen train cancer research models, including single-cell RNA transformers with models up to 3B parameters running in BF16/FP32. It reports a best TLS prediction result of 0.94 ROC-AUC without spatial coordinates and says Nebius helped integrate AI into Compugen’s Unigen discovery pipeline.
Cloud for Scientific AI and Healthcare13 Oct 2026 · nebius.com · Web pageNebius describes helping Compugen train cancer research models, including single-cell RNA transformers with models up to 3B parameters running in BF16/FP32. It reports a best TLS prediction result of 0.94 ROC-AUC without spatial coordinates and says Nebius helped integrate AI into Compugen’s Unigen discovery pipeline.
Nebius Customer story
27 Sept 2026 · Nebius · Web page
Nebius describes Compugen’s collaboration expanding from an educational partnership at Nebius Academy into full-scale training on Nebius AI Cloud. The work produced a research model for tertiary lymphoid structure prediction using scGPT, scGPT-Spatial and Nicheformer.
Nebius Customer story27 Sept 2026 · Nebius · Web pageNebius describes Compugen’s collaboration expanding from an educational partnership at Nebius Academy into full-scale training on Nebius AI Cloud. The work produced a research model for tertiary lymphoid structure prediction using scGPT, scGPT-Spatial and Nicheformer.
#aiinhealthcare #cancerresearch #nebiusacademy | Nebius Academy Israel
13 Oct 2026 · www.linkedin.com · Web page
Nebius Academy Israel names Compugen as its industry partner for a graduate TLS classification project. Graduates worked with Compugen specialists using scGPT, scGPT-Spatial and Nicheformer across 14 cohorts and hundreds of thousands of cells. Nebius supplied the Academy project team with a special GPU Cloud grant.
#aiinhealthcare #cancerresearch #nebiusacademy | Nebius Academy Israel13 Oct 2026 · www.linkedin.com · Web pageNebius Academy Israel names Compugen as its industry partner for a graduate TLS classification project. Graduates worked with Compugen specialists using scGPT, scGPT-Spatial and Nicheformer across 14 cohorts and hundreds of thousands of cells. Nebius supplied the Academy project team with a special GPU Cloud grant.
Compugen is a clinical-stage biotechnology company developing cancer immunotherapies. Its Unigen platform uses AI, machine learning and biological data to identify drug targets and guide antibody development. Its pipeline includes COM701, an anti-PVRIG antibody being studied as maintenance therapy for relapsed ovarian cancer; rilvegostomig, a PD-1/TIGIT bispecific antibody developed by AstraZeneca using a component derived from Compugen’s COM902; and GS-0321, an antibody licensed to Gilead and undergoing a Phase 1 trial. Its programs target cancer patients and support pharmaceutical partners’ drug development.
At its October 5 platform presentation, Compugen introduced UniGenX, an expansion of its AI discovery engine focused on tumor-associated antigens. It combines multiple biological data types to prioritize targets for cancer therapies, with candidates undergoing further experimental validation.
Compugen reports second-quarter revenue of $2.6 million and $7 million net loss
Compugen reported second-quarter revenue of approximately $2.6 million, up from $1.3 million a year earlier, and a net loss of $7 million. Cash and investments totaled $125.3 million at June 30, with operations expected to be funded into 2029. Interim progression-free survival data from its COM701 MAIA-ovarian trial remained expected by the first quarter of 2027.
