Google Shares AI Advances in Health, Weather and Planetary Prediction
On September 15 2026 Google highlighted recent tools including the openly available AlphaGenome Atlas for genetic insights, the more accurate WeatherNext 3 weather model…
More freedom.
More control.
New AI releases, creative tools and what you can actually download, change and use.
118 stories
On September 15 2026 Google highlighted recent tools including the openly available AlphaGenome Atlas for genetic insights, the more accurate WeatherNext 3 weather model…
Players in Steering Arena found that unreadable token strings shifted Olmo 3-32B's internal activations toward prosocial responses far more effectively than plain-English…
Ai2's new open infrastructure scales mixture-of-experts training to over a trillion parameters with minimal throughput loss, letting researchers and smaller teams build larger…
Reflection launched Beam, a 501-billion-parameter mixture-of-experts model with 23B active parameters per token, 1M-token context window, and Apache 2.0 license. Users can…
Mistral announced a €3 billion Series D funding round at over €21 billion valuation, led by Samsung Electronics. The capital will scale frontier research, compute capacity…
Mistral released a public preview of its 1-trillion-parameter multimodal model Mistral Large 4 on October 6 2026. It leads open-weight models on cybersecurity benchmarks where…
Black Forest Labs released FLUX 3 Image Klein, a 5.6B parameter open-weights model that runs on consumer GPUs for text-to-image generation, precise editing, and multi-reference…
As of October 16 2026 Reflection's first open-weight Beam model with 501B total parameters is available via Ollama Cloud with zero data retention. Users gain immediate cloud…
An arXiv preprint released October 16 2026 finds that a small language model fine-tuned with LoRA in 2.5 hours on an Apple M5 Pro reaches 69.9 percent severity-tier accuracy…
An October 19 2026 arXiv preprint presents a large-scale study of parameter-efficient fine-tuning design choices on three MoE model backbones, evaluating module selection…
A September 2026 arXiv preprint applies sparse crosscoders and a new swap readout technique to examine what on-policy distillation actually changes inside student LLMs during…
A September 2026 arXiv preprint hypothesizes that LLM layers divide labor across conceptualization, reasoning and textualization stages and introduces Layer-Informed…