Microsoft finished pre-training MAI-Base-2, a 2.8T-parameter model it says is 4–12x more efficient than DeepSeek’s best base model. Not usable yet.
Category: iRaluca
Swiss AI’s Apertus 1.5 adds images, audio and a 262k context. Now every training stage, dataset and script behind it is public too.
Meta’s open-weight Muse Spark Open barely jailbreaks in its own tests, but refuses about 1 in 6 harmless prompts. And fine-tuning is allowed.
The EU’s AI Office is probing OpenAI and Anthropic over models that broke out of testing. The catch: the AI Act exempts testing.
Liquid AI’s open-weight LFM2.5-24B-A2B runs agents on laptops with 2B active parameters. Strong at tools, weaker on facts, by its own admission.
Both labs said outside evaluators should control their findings. Reuters says that in private, both sought a say over what could be published.
Anthropic tested whether small open models could copy themselves across cheap gadgets. One big open model pulled it off, in a simplified test.
Meta’s Muse Wonder promises an agent that remembers you between chats. The model card’s numbers are in; the memory controls aren’t.
The FT reports Nvidia is in early talks to buy or deepen its stake in Reflection AI, whose promised Apache 2.0 Beam weights still aren’t out.
JetBrains’ open Mellum2.1 uses 2.5B active parameters. New training alone lifted its SWE-bench Verified score from 2% to 47%, self-reported.