Microsoft's new Canvas workspace runs on MAI Image 2.7 Flash. You can sketch, select and edit parts of an image in one place, at about a cent per image through the API.
Linger, a new open-source personal agent, runs on consumer PCs. It handles memory, agent teams and email triage locally using a MoE engine that fits a 125B model onto a 12 GB GPU.
A NeurIPS paper looks inside looped language models and finds that each pass through the shared block repeats the same processing stages a normal Transformer runs once. Whether those stages hold steady depends on a few design choices.
Monash and German researchers prune the Hydra and Quant time series classifiers, cutting inference energy by up to 70% with only a small loss in accuracy, and show that hardware choice can change which model looks best.
Kling Image 4.0 jumps more than 225 Elo over its predecessor and lands in the top 7 for both generation and editing. It adds native 8K output and better text rendering, and charges about a third of the price of the OpenAI models ranked above it.
A revised MIT/Oxford paper shows that tuning AI classifiers for accuracy can quadruple worst-case liability, and proposes insurance contracts that force safer decision thresholds.
Prime Intellect's Model Factory puts pre-training, fine-tuning, large-scale RL, sandboxes, inference and trace collection on one stack, so teams can own their models' full improvement loop.
Google's new middle-tier Gemini 4 Neon is live in Cursor and the Gemini API. On CursorBench it beats Carbon, Google's flagship, at a lower cost per task, which makes it a strong candidate for your default coding agent.
Anthropic priced its upsized IPO at $83 a share, above its range. The $81.3 billion raise is the largest on record and values the Claude maker at about $1.5 trillion.
Meituan's open-weight talking-avatar model now holds a consistent face for more than 10 minutes, generates in 8 steps, and placed second only to OmniHuman-2.0 in Meituan's own human-preference vote.
Google's new Gemini 4 Neon sits between Carbon and Hydrogen. It scores 61 on the Artificial Analysis Intelligence Index, just above Carbon, at $3/$15 per million tokens.
Pika's Enact research preview streams AI video and sound in under-a-second chunks, so you can drag actors, talk to them and move the camera while the scene keeps playing.
Paweł Huryn's free, MIT-licensed product-engineering pack gives coding agents 45 skills that take a product from first idea to paying users, and no session is allowed to review its own code.
Genie 4 is now live in Project Genie for Google AI Ultra subscribers. You can edit a world mid-session by typing, sound is generated with the visuals, and a Gemini agent helps build and remix worlds.
Imbue has open-sourced Sculptor under MIT. The desktop app runs Claude Code, Pi and other coding agents in parallel, so your workflow isn't tied to one AI lab.
MI-Pruner decides which image tokens to drop by measuring how much each one shares with the text prompt. It never reads attention maps, so it can be added to off-the-shelf multimodal models without changing them.
Parallel retuned its search index for FDA labels, trial registries and PubMed papers. In its own tests on four healthcare benchmarks, it was the most accurate search engine and often the cheapest per correct answer.