Mistral opens a public-preview API for its 1-trillion-parameter multimodal Mistral Large 4, plans to release open weights by October 2026, and is red-teaming a less-moderated cyber-capable version with vetted partners and state authorities.
Mistral AI launches its 1-trillion-parameter multimodal Mistral Large 4, initially restricting access to a public guardrail endpoint before planning to release the model’s weights in three weeks following safety testing.
Scale Labs open-sources AgentEnv, a modular multi-agent reinforcement-learning framework that runs the same task across Claude Code, OpenAI, and custom agents using Environment and Agent Cards, with 48 built-in step types and four sandbox options.
Google launches EmbeddingGemma 2, a 740-million-parameter Apache 2.0 model that embeds text, code, images, video and audio in one on-device space, supporting 8K-token context and using as little as 191MB of active RAM on a Pixel 11 Pro when quantized.
Aleph Alpha releases Kolibri, a sovereign German-English open-weight AI model for mission-critical government and industry use, combining 78 billion total parameters with roughly 3 billion active per token and training data screened against a 4.5-million-URL blocklist.
Reflection AI unveils Beam, an open-weight coding and agent model it says matches Z.ai’s GLM-5.2 reasoning and coding performance with a fraction of the compute, with Apache 2.0-licensed weights due this month.
DeepSeek releases DeepGEMM Ascend, using DeepJIT runtime kernel compilation to avoid CUDA builds at installation and a fused Mega MoE kernel that overlaps NVLink transfers with tensor-core computation.
Reflection unveils Beam, a 501-billion-parameter Mixture-of-Experts reasoning model with 23 billion active parameters, and says it will release Apache 2.0-licensed weights and developer tools later this month.
Cognition adds persistent Memory and daily “Dreaming” to Devin, enabling the coding agent to retain each user’s preferences, corrections and project lessons in a Git-backed Markdown repository across sessions, while open-sourcing its agent-memory standard.
Gatewai launches the beta of Framefields, an Apache-2.0 Node.js package that renders code-first video directly via WebGPU, Dawn, Metal or Vulkan instead of headless Chromium, while offering on-device ONNX vision tools for segmentation, pose tracking and person mattes.