Source-checked AI news, every morning. Each day brings six articles and one digest on new models from labs worldwide, research, essays, local models and agents, and prompting techniques. Every article ends with a verification table that links the primary sources.
How we work · Archive · RSS feed · For AI assistants
Other editions: Slovenčina · Nederlands · Français · Deutsch · Español · 简体中文
The small Claude model adds adaptive thinking and a two-tier token price. Anthropic positions it for summaries, browser work and narrowly scoped subagents.
The 3.1-billion-parameter d1-3B and experimental 600-million-parameter d1-omni answer structured questions in one forward pass, with weights available for local deployment.
A preprint finds that a few fixed words at the start of an answer can recover much of the reasoning performance usually associated with reinforcement learning.
A preprint traces how models that update their weights from their own generated text can enter a damaging feedback loop over long inference streams.
The Apache-licensed project keeps Python and model code on a client machine while tensors and CUDA operations run on a remote NVIDIA GPU.
InnovationEval gave two frontier agents code, metrics and large GPU budgets, then asked them to rediscover a recent training method. Neither produced a comparable innovation.
Open models, research preprints, agent projects, technical essays and community measurements: the rest of today's AI news with direct sources.
Google's 740-million-parameter embedding model maps text, code, images, video and audio into one vector space, with modular encoders designed for local devices.
Mistral has opened API access to its one-trillion-parameter multimodal model and set 27 October for the open-weight release, leaving architecture and benchmark claims provisional.
A preprint finds an early object-matching circuit and a later relation-matching circuit in vision-language models, then links the latter to performance on ARC-AGI-1.
The Apache-licensed project publishes its accelerator design, compiler, simulator and host tools, with measured throughput showing a small, memory-bound system rather than a GPU replacement.
A preregistered experiment compares nine Claude Code skills with equally long neutral instructions, finding two cheaper than placebo, one worse and six statistically indistinguishable.