Auto Added by WPeMatico
MiniMax officially released MiniMax M3 on June 1, 2026. The model introduces MSA (MiniMax Sparse Attention), a new sparse attention architecture that gives M3 a 1M-token context window. M3 also supports image and video input and desktop computer operation natively. The API is live now. MiniMax M3 is available today via MiniMax Code, the MiniMax […]
Hermes Agent already remembers across sessions. The open-source agent from Nous Research ships with curated memory files and full-text session search. But a new community project argues that built-in memory is too shallow for serious work. A new library named ‘Memory OS‘ has been released under an MIT license by a developer (ClaudioDrews). It stacks
Meet Memory OS: A 6-Layer Open-Source Memory Stack Built on Top of Hermes Agent Read More »
The Transformer’s attention mechanism has barely changed since 2017. Most efficiency work has tried to replace softmax attention outright. A new paper takes a different route. It keeps softmax attention and bolts on a correction branch. A team of researchers from Northwestern University, Tilde Research, and University of Washington introduce a parameterized Local Linear Attention
Trajectory’s concurrent multi-LoRA stack reports a 2.81× experiment-throughput gain over single-tenant RL, with all code in the NovaSky-AI/SkyRL GitHub repository. Most language models improve in discontinuous jumps. A team collects data, trains, and ships a new version. This takes months and produces remarkable or catastrophic behavior for users. Trajectory wants to replace that cycle with
Genesis AI released Genesis World 1.0. The platform consists of four components: the Genesis World physics engine, Nyx (a real-time path-traced renderer), Quadrants (a Python-to-GPU compiler), and a simulation interface. It is designed to accelerate robotics foundation model development through simulation-based evaluation. Robotics model development has two bottlenecks: data and iteration speed. The field has
Nous Research’s open-source Hermes Agent now ships a Tool Search feature. It directly addresses a growing bottleneck in AI agent systems: too many MCP tools filling up the context window. In this explainer article, we will breaks down what Tool Search does, how it works, and when to use it. The Problem: MCP Tools Are
Knowledge distillation (KD) transfers “dark knowledge” from a large teacher model to a smaller student. The student learns from the teacher’s full output probability distribution over tokens, not just correct answers. This is done via per-position Kullback–Leibler (KL) divergence over next-token probability distributions. This formulation requires a shared tokenizer. A practitioner committed to Llama-3.2-1B cannot
StepFun today released Step 3.7 Flash, a multimodal Mixture-of-Experts model targeting agentic use cases. It adds native vision input and improved tool-use reliability over Step 3.5 Flash. What is Step 3.7 Flash? Step 3.7 Flash is a 198B-parameter sparse Mixture-of-Experts (MoE) vision-language model. It pairs a 196B-parameter language backbone with a 1.8B-parameter vision encoder (ViT)
GPU communication overhead is a measurable bottleneck in production AI workloads. According to data cited by the mKernel project, communication can consume 43.6% of the forward pass and 32% of end-to-end training time. Across popular Mixture-of-Experts (MoE) models, inter-device communication can account for up to 47% of total execution time. Researchers from UC Berkeley’s UCCL
Meet mKernel: A Multi-GPU, Multi-Node Fused Kernel Library for GPU-Driven Communication Read More »
Liquid AI just shipped LFM2.5-8B-A1B. It is an on-device Mixture-of-Experts (MoE) model built for tool calling. The model holds 8.3B total parameters but activates only 1.5B per token. That sparsity is what lets it run on consumer hardware. The release follows LFM2-8B-A1B, which Liquid AI team published earlier. LFM2.5 is a new family of hybrid
