M*: New Serving System for Multimodal AI Models Shows Performance Improvements
Researchers have introduced M*, a modular serving system designed to efficiently run composite AI models that combine vision, language, audio, and other components. Existing serving frameworks were built around simpler, single-modality assumptions and struggle with the architectural complexity of modern multimodal systems. M* addresses this gap by representing models as dataflow graphs, enabling flexible deployment and optimization across distributed hardware.
A team of researchers from multiple institutions has published a preprint on arXiv introducing M*, a universal serving system for composite AI models. As AI architectures increasingly combine diverse components—such as vision encoders, language backbones, diffusion heads, audio codecs, and action generators—existing frameworks like vLLM have become ill-suited to handle their structural complexity. M* represents models as 'Walk Graphs,' a modular abstraction that supports arbitrary composition of components, flexible physical cluster placement, and model-agnostic runtime optimizations. In benchmark evaluations, M* achieved on average 20% lower end-to-end latency than vLLM-Omni on text-to-image workloads using the BAGEL model, up to 2.9x lower real-time factor and 2.7x higher throughput on text-to-speech workloads with Qwen3-Omni, and up to 12.5x better performance than the V-JEPA 2-AC baseline for robotic planning tasks. The authors argue that M* reduces developer effort while enabling more efficient serving of increasingly complex multimodal systems.
What's missing
As a preprint, M* has not yet undergone peer review. The benchmarks are self-reported by the authors, and independent third-party replication has not been conducted. The paper does not fully characterize performance trade-offs under diverse real-world production conditions, varying cluster sizes, or with models outside the tested set. Scalability limits and failure modes of the Walk Graph abstraction remain open questions.
What different sources said
- arXiv cs.AICenter
M*: A Modular, Extensible, Serving System for Multimodal Models
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.