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 bidirectional encoders that processes text and raw 32×32 image patches through a single Transformer, with no vision tower or decoder. NeoMME-Retriever-260M scores 0.523 nDCG@10 on ViDoRe v3, within 0.002 of 3.75B-parameter ColQwen2.5. Checkpoints ship under Apache 2.0 with Hugging Face support; the 260M model indexes 51.3 pages per second on an NVIDIA L40S. Text retrieval on BEIR-15 trails LateOn.