Models & Labs
NVIDIA NVFP4 and SoL-Pi Token Efficiency
Lev Selectorhigh confidence
Why it matters
- →Reduces inference costs and energy use.
- →Improves scalability for large language models.
- →Strengthens NVIDIA's role in AI infrastructure.
NVIDIA has unveiled NVFP4, a new 4-bit quantization format designed to accelerate AI inference while maintaining model accuracy. Additionally, the company introduced SoL-Pi, a technology that reduces token usage by half for equivalent performance. These innovations aim to lower computational costs and energy consumption for large-scale AI deployments. The updates position NVIDIA as a key enabler of efficient AI hardware and software stacks.
Read originalMore from Lev Selector
© Lev SelectorMarket & Regulationbusiness
OpenAI Eyes $1.5 Trillion Valuation
Reports indicate OpenAI is targeting a valuation of $1.5 trillion in its next funding round, reflecting massive investor confidence.
Lev Selector
© Lev SelectorMarket & Regulationbusiness
Cohere and Aleph Alpha Announce $20B Merger
AI infrastructure firms Cohere and Aleph Alpha have announced a merger valued at $20 billion, creating a major player in the enterprise AI market.
Lev Selector
© Lev SelectorAgentsagents
AWS Strands Agent Harness Released
AWS launched Strands Agent Harness, a tool to manage AI agents, with reports showing up to 5x cost differences based on harness choice.
Lev Selector
More in Models & Labs
Models & Labsother
llama.cpp b11193 adds Snapdragon Hexagon and ROCm 10
This release quietly expands llama.cpp's hardware support to include Qualcomm's Hexagon NPU on Linux arm64, a significant step for local inference on Snapdragon devices. It also updates CUDA builds to version 13.4 and introduces ROCm 10.0 binaries, keeping the project aligned with the latest NVIDIA and AMD driver ecosystems. KleidiAI on Apple Silicon is temporarily disabled in this build, likely due to stability checks rather than a feature rollback. For developers targeting edge AI or diverse GPU stacks, this update ensures broader compatibility without requiring custom compilation.
llama.cpp Releases
© TechCrunch AIModels & Labsresearch
GPT-6 Astra and Claude Opus 5 crack unsolved Enigma codes
OpenAI’s GPT-6 Astra and Anthropic’s Claude Opus 5 have independently broken long-standing Enigma ciphers that human cryptanalysts failed to solve for nearly two decades. This isn't just pattern matching; the models performed archival research, built simulators, and leveraged contextual clues to recover plaintext from messages dating back to 2005. The achievement demonstrates a leap in autonomous reasoning and tool use, effectively turning LLMs into professional researchers capable of multi-step problem solving that previously required weeks of human effort. It marks a significant shift in what we expect from frontier models beyond simple text generation.
TechCrunch AI
© The Verge AIModels & Labsagents
Microsoft Unifies Copilot into Single 'Super App'
Microsoft is consolidating its fragmented AI tools into a single interface that merges chat, coding, and autonomous agents under one roof. The new app bundles the rebranded Autopilot agent with GitHub Copilot’s code generation capabilities, targeting enterprise workflows rather than consumer competition. By introducing usage-based billing for these agentic features, Microsoft is shifting how businesses pay for AI utility beyond flat subscriptions. This move signals a pivot from standalone chatbots to an integrated operating system for work, aiming to replicate the dominance Office held in the PC era.
The Verge AI