H Company Releases NeoMME: A Family of 260M and 800M Single-Tower Multimodal Encoders That Drop the Vision Tower and Causal Decoder

H Company released NeoMME, a family of 260M and 800M single-tower multimodal encoders that process multilingual text and raw 32×32 image patches in one Transformer without a pretrained vision tower or causal decoder. On the ViDoRe v3 benchmark, the 260M model achieved 0.523 nDCG@10, with 255× index compression and 51.3 pages per second indexing throughput on one L40S GPU.

We look at NeoMME, a family of 260M and 800M bidirectional encoders from H Company. Unlike ColPali-style retrievers, it processes multilingual text tokens and raw 32×32 image patches in a single Transformer, with no pretrained vision tower and no causal decoder. We cover the masked discrete-diffusion pretraining objective, the dual dense and late-interaction retrieval heads, and the ViDoRe v3 results where the 260M model reaches 0.523 nDCG@10. We also break down the 255× index compression, the 51.3 pages per second indexing throughput on one L40S, and the text-retrieval gaps the authors acknowledge. The post H Company Releases NeoMME: A Family of 260M and 800M Single-Tower Multimodal Encoders That Drop the Vision Tower and Causal Decoder https://www.marktechpost.com/2026/09/06/h-company-releases-neomme-a-family-of-260m-and-800m-single-tower-multimodal-encoders-that-drop-the-vision-tower-and-causal-decoder/ appeared first on MarkTechPost https://www.marktechpost.com .