All AI Pulse updates tagged "Product Update".
14 signalsGoogle CEO Sundar Pichai has released Google's latest Q2 earnings report: Alphabet's revenue grew 24% year-over-year, Google Cloud surged 82%, and Gemini's monthly active users reached 950 million. The model API processes approximately 22 billion tokens per mi
For over two years, the largest open code dataset was The Stack v2… until today. The Stack v3 is out: the largest open code dataset ever released: 114 TB, 770 languages, 224M repositories, ~5T tokens of deduplicated and filtered source code. Fully open, no res
Most AI SDKs are a wrapper around one engine on one platform. We built the opposite. 7 SDKs. Swift, Kotlin, Flutter, React Native, Web, Electron, rcli. Python. All of them are thin skins over one C++ core. 33 protobuf contracts. One capability registry. Under
It looks like FLUX 3 is about to be released. Yesterday, Black Forest Labs briefly put up a teaser. They say it supports image, video, audio, and motion generation—it's a full-modality generative model. Among its features, a single video can be generated in on
Claude Code 2.1.193 has been released. 15 CLI changes Highlights: • Bash and PowerShell commands now go through the auto-mode classifier to enforce uniform shell safety checks • Auto-mode records denial reasons in transcripts, denial toast, and /permissions fo
Customer by customer it is gonna take an entire year to release the model. The fact that government is micromanaging, the release is insane.
Claude Code 2.1.181 has been released. 39 CLI changes, 1 system prompt change Highlights: • Prompts now include Bun HMR warnings that surface hot-reload diagnostics, cluttering or exposing build info • Added /config key=value to set any setting inline from pro
v2.42.0
Anthropic built a model so powerful they gave it to the government before the public ever saw it. Then they added safeguards. Then they released it anyway. It's called Claude Fable 5. Mythos class. The same tier that Anthropic internally flagged as too capable
Released last week, and already more than 4M downloads on HuggingFace alone This makes Gemma 4 12B the most popular encoderfree VLM by a large margin. In addition to being the first-ever general purpose LLM with encoderfree audio input!
NVIDIA has just released a 0.6B speech recognition model called Nemotron-3.5-ASR supports over 40 languages, real-time streaming output—no GPU needed, runs purely on CPU speed is 2.5 times that of the official Nemo runtime, with completely identical recognitio
Meet DiffusionGemma! An experimental open model that explores a fast approach to text generation, released under an Apache 2.0 license. Moving beyond sequential, token-by-token processes to generate entire blocks of text simultaneously. Here’s what’s new with
a guy opened the Fable 5 release page and found the part everyone will actually remember not the benchmark chart. the examples. > it beat Pokémon using only screenshots > played Factorio autonomously > and Stripe used it on a 50M-line code migration that is th
v0.109.1