Tool Reviews
This release uses sparse autoencoders to show Qwen3 MoE experts have private, fine-grained specializations, releasing code and an expert explorer.
Nov 22, 2026
Tool Reviews
Reflection's Beam is a 501 Billion Apache-2.0 open LLM with 1 Million token context for coding, agents, and GPU deployment.
Nov 19, 2026
Tool Reviews
MLExpert, a 35-billion Mixture-of-Experts agentic model, uses execution feedback to lift MLE-Bench Lite medal rate to 72.7%.
Nov 19, 2026
Tutorials
LFM2.5-24B-A2B is a local reasoning agent with two billion active parameters, fast Apple M5 Max inference, and a 131,000-token context.
Oct 21, 2026
Tool Reviews
StepFun launched Step 5 Preview, a flagship sparse Mixture-of-Experts model with 600B total parameters (27B active per token) and a 1M-token context window, targeting agentic workloads like coding, re…
Sep 22, 2026