DISPATCH // AI NEWS
Latest AI News
Signal over noise. Concise, curated news on AI models, tools, and prompt engineering for people who ship.
ModelsAug 14, 20262 min read
Writer Launches GLM-5.2-Based Model With a Token-Cost Harness Built In
Writer has released a new model built on Z.ai's open-source GLM-5.2, paired with a cost-containment harness designed to make deployment economics predictable. For teams shipping LLM features at scale, this is a direct answer to runaway token spend.
ModelsAug 5, 20262 min read
LFM2.5-2.6B Lands: Liquid AI's Tiny Model Makes Local Agents Practical
Liquid AI released LFM2.5-2.6B, a sub-3B parameter model optimized for on-device agent deployment. Here is what the efficiency gains mean for teams building local-first LLM pipelines.
ModelsJul 28, 20262 min read
Kimi K3 Weights Drop: 2.8 Trillion Parameters, Open for Self-Hosting
Moonshot AI has released the full weights for Kimi K3, a 2.8 trillion parameter model, on Hugging Face. At 1.56TB, it is now the largest openly available model weights for teams willing to run their own inference.
ModelsJul 10, 20262 min read
GPT-5.6 Family Launches: Luna, Terra, and Sol Hit General Availability
OpenAI's GPT-5.6 family is live in three tiers, with pricing from $1/$6 to $5/$30 per million tokens. Here is what the new lineup means for teams choosing models right now.
ModelsJun 24, 20262 min read
Qwen-AgentWorld-35B Is a World Model for Agents, Not Another Chat Model
Qwen released AgentWorld-35B-A3B, a MoE model trained to simulate what environments return after agent actions, not to chat or plan. This reframes how agent pipelines can be built and tested.
ModelsJun 17, 20262 min read
GLM-5.2 Becomes the New Open-Weights Leader, Beating GPT-5.5 on Agentic Knowledge Work
Z.ai's GLM-5.2, a 753B-parameter MoE under the MIT license, is now the top open-weights model on the Artificial Analysis Intelligence Index and scored above GPT-5.5 on the new AA-Briefcase agentic eval. The frontier-grade option you can self-host just shifted.
ModelsJun 12, 20262 min read
Gemma 4 12B Drops the Vision Encoder for Simpler Multimodal Deployment
Google DeepMind's Gemma 4 12B fuses vision and language into one encoder-free architecture, cutting deployment complexity for self-hosted and edge inference.