San Francisco Compute, which provides a marketplace for AI computing capacity, raised a $40M Series A led by DCVC and Wing Venture Capital at a $300M valuation
Yuliya Chernova / Wall Street Journal :
Gemini co-lead Oriol Vinyals says Gemini 3's gains come from better pre-training and post-training, contradicting the idea that pre-training gains are falling
which we discussed in our NeurIPS '25 talk with @ilyasut and @quocleix—the team delivered a drastic jump. The delta between 2.5 and 3.0 is [image] Andrej Karpathy / @karpathy : I p...
Gemini 3 hands-on: a fundamental improvement on daily use, extremely fast, Antigravity IDE is a powerful launch product, and its personality is terse and direct
Gemini 3 is a fundamental improvement on daily use, not just on benchmarks. It feels more consistent and less “spiky” than previous models.
Soumith Chintala, who co-created the PyTorch ML framework at Meta and left the company earlier this month, joins Mira Murati's Thinking Machines Lab
Pranav Dixit / Business Insider :
Extropic, which says its chips using probabilistic bits can be 10,000x more energy efficient than current AI chips, shares its first chip with some AI labs
A startup hopes to challenge Nvidia, AMD, and Intel with a chip that wrangles probabilities rather than 1s and 0s.
OpenAI completes its recapitalization, “simplifying” its structure; OpenAI Foundation now has equity valued at ~$130B and still controls the OpenAI for-profit
OpenAI has completed its recapitalization, simplifying its corporate structure. The nonprofit remains in control …
Microsoft now holds a ~$135B investment in OpenAI Group PBC, or a ~27% stake, down from 32.5%; OpenAI commits to purchase an additional $250B in Azure services
Since 2019, Microsoft and OpenAI have shared a vision to advance artificial intelligence responsibly and make its benefits broadly accessible.
Andrej Karpathy unveils nanochat, a full-stack training and inference implementation of an LLM in a single, dependency-minimal codebase, deployable in 4 hours
It provides a full ChatGPT-style LLM, including training, inference and a web Ui … X: Clem / @clementdelangue : Am I wrong in sensing a paradigm shift in AI? Feels like we're movin...
Google announces Gemma 3 270M, a compact model designed for task-specific fine-tuning with strong capabilities in instruction following and text structuring
ai.google.dev/gemma/docs/c... Tim Duffy / @timfduffy.com : Google just released a 270M parameter Gemma model. As a tiny model lover I'm excited. Models in this size class are usu...
A study by Meta researchers suggests that training LLMs to predict multiple tokens at once, instead of just the next token, results in better and faster models
LLM approach to predict multiple tokens KAN: Kolmogorov-Arnold Networks —"promising alternatives to Multi-Layer Perceptrons" [image] Ethan / @ethan_smith_20 : it was only briefly t...
A look at Databricks' new open-source model DBRX, an LLM that cost ~$10M to train over several months and, Databricks says, outshines Llama 2, Mixtral, and Grok
Startup Databricks just released DBRX, the most powerful open source large language model yet—eclipsing Meta's Llama 2.
A look at Databricks' new open-source model DBRX, an LLM that cost ~$10M to train over several months and, Databricks says, outshines Llama 2, Mixtral, and Grok
Startup Databricks just released DBRX, the most powerful open source large language model yet—eclipsing Meta's Llama 2.
A look at Databricks' new open-source model DBRX, an LLM that cost ~$10M to train over several months and, Databricks says, outshines Llama 2, Mixtral, and Grok
Startup Databricks just released DBRX, the most powerful open source large language model yet—eclipsing Meta's Llama 2.