Tag
4 articles
Mistral AI has released Mistral Large 4, a 1.05 trillion parameter multimodal MoE model with native image input and a 1 million token context window. The API is live, with open weights to follow in October 2026.
Reflection AI introduces Beam, a 501B sparse Mixture-of-Experts model with 23B active parameters designed for coding and agentic workloads, offering performance comparable to GLM-5.2 with significantly reduced compute requirements.
Learn how to load, configure, and run inference on Aleph Alpha's Kolibri MoE model with only 3.46B active parameters out of 78.1B total parameters.
Z.ai has released GLM-5.3-Flash, a 320B-parameter, 18B-active MoE model with a 1M-token context window and native multimodal capabilities. It features a 3x reduction in attention compute and 4.4x in KV cache usage compared to previous versions.