Ai Models

Reka AI has unveiled Rho-1, a 19-billion-parameter omni-model that processes and generates text, images, video, and robot control actions in a single neural network. Trained on 320 H100 GPUs in about three months, it uses a fraction of the compute of today's top models.

AI Models • The Decoder

Mistral has launched Mistral Large 4, a 1 trillion-parameter multimodal model with 49 billion active parameters, in public preview. The company claims it pushes the frontier of open-weight performance, with weights expected to be released at the end of the month.

AI Models • Hacker News

Microsoft published a bearish AI economic forecast from Nobel economist Daron Acemoglu, predicting only 1.5% GDP growth over a decade and at most 5% of jobs replaced. Acemoglu argues that larger models won't significantly change this outlook, and that practical applications are what's missing.

AI Economy • The Decoder

Google is set to remove free access to its Gemini Flash and Pro models starting October 9th, limiting free users to the Flash Lite model. Access to the standard Flash model will require a $4.99/month Google AI Plus subscription, which is also being updated.

AI Products • The Verge AI

A new lightweight decision model called PolicyLM-1.7B, released with open weights by Musubi, is designed for real-time content moderation. The model aims to change how AI decision models are used in this space.

AI Models • TechCrunch AI

Productivity

Use AI agents to iteratively improve your own prompts or models, but be prepared for unexpected strategies that may emerge.

Microsoft revealed the specs and price for its Surface Laptop Ultra, AI PCs that run on Nvidia chips designed to run AI models and agents. The company also showed off an upgrade to its Copilot AI system that will give it access to local files and the ability to take actions across the OS, part of a concept called 'Hybrid Intelligence'.

Hardware • TechCrunch AI

OpenAI has launched a new Decisions API that classifies text and images about ten times faster than its Responses API, returning yes/no probabilities, category picks, or scale ratings. The company also cut its paid API tiers from five to three.

AI Models • The Decoder

Mistral has released a new model, reportedly nicknamed 'Le Chonk', which the company claims can rival top closed models while remaining open-weight. The model is positioned as a challenger to leading proprietary AI systems.

AI Models • Ars Technica AI

Fireship breaks down the latest moves in the open-weight AI race, including Mistral, Reflection AI, and Moonshot. It highlights how a French model outperformed a $6.3 billion open-weight model, offering insights into the competitive landscape and technical advancements in open-source AI.

AI Models • Fireship

A new open-source decision model called Strands Decider 2B has been released, optimized for fast experimentation and local development. It is part of a new class of 'decision models' or 'system one models' that have gained attention since TypeSafe AI's launch of Jev.

AI Models • Hacker News

Anthropic is expanding its Cyber Verification Program, giving more security professionals access to Claude models with fewer safety restrictions for penetration testing, malware analysis, and vulnerability research. The company says partners in its predecessor program have seen success.

AI Safety • The Decoder