JetBrains Mellum2.1 is an open 12B MoE coding agent model with 2.5B active parameters and 47.0% SWE-bench Verified.

Claude Haiku 5.5 brings 1M context, adjustable effort, and $0.10 input pricing to high-volume subagent and browser workloads.

NVIDIA's PivotOPD trains multi-turn agents to recover from pivotal mistakes and beats 13 baselines on ALFWorld, WebShop, ...

Architect's Liquid Inference auctions every LLM request across competing providers, locking a max price before the first ...

Unsloth details how Studio scans model code, blocks flagged weights, inspects packages and sandboxes tools before anything ...

Liquid AI has released Open d1, two open-weight multimodal models in its d1 decision model family. d1-3B reads text and ...

Mistral AI released Mistral Large 4, a 1.05T parameter open-weight multimodal MoE with 49B active parameters, 1M context.

Reka Rho-1: a 19B omni-reasoning model unifying text, image, video generation and robot actions inside one shared network.

Google DeepMind's EmbeddingGemma 2 embeds text, code, images, video and audio in one 740M model for on-device RAG.

JEPA-Anything adds Orthogonal Predictive Factorization to JEPA world models, improving all 10 matched dynamics tasks across 7 ...

Reflection AI's Beam is a 501B open-weight MoE with 23B active parameters, 1M context, and Apache 2.0 weights.

Meta open-sources Rebalancer, the Apache 2.0 C++ solver behind 40 million daily assignment problems across its global infrastructure.