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.
Tag: open source
Open encoder family UltraBERT trains with two masking ratios at once to be good at both classifying text and finding it. Apache 2.0.
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.
Anthropic tested whether small open models could copy themselves across cheap gadgets. One big open model pulled it off, in a simplified test.
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.
Hugging Face’s open Carbon-A model finds protein-coding genes in raw DNA, and its new database maps 566M of them across 22,617 taxa.
Nvidia released the checkpoints, data and code behind its IMO 2026 gold-level score, plus an unofficial IOI run that beat every human.
Liquid AI’s open-weight d1 models answer yes/no, pick-one and score questions about text, images and speech in one pass, without writing text.
Google released EmbeddingGemma 2, a 740M Apache 2.0 model that maps text, code, images, video and audio into one searchable space.