Models & Labs
Salesforce and Nvidia Launch Koa AI Model
TechCrunch AIhigh confidence
Why it matters
- →Koa offers a task-specific AI model that enhances privacy by not using customer data.
- →It provides a cost-effective alternative to traditional AI models by being token efficient.
- →This development signals a shift towards enterprise-focused AI solutions, challenging existing AI lab models.
Salesforce has unveiled Koa, a new AI reasoning model developed in collaboration with Nvidia, at its Dreamforce conference. Built on Nvidia's Nemotron, Koa is designed to excel in sales and customer support tasks without using actual customer data, ensuring privacy and cost efficiency. This model represents a shift towards enterprise-specific AI solutions, diverging from the broader capabilities offered by frontier labs. Koa will be integrated into Salesforce's Agentforce platform, providing a tailored alternative to models like Claude and ChatGPT.
Read originalMore from TechCrunch AI
© TechCrunch AIMarket & Regulationbusiness
Nvidia's CEO Opposes New AI Regulations
Nvidia CEO Jensen Huang argues against new AI regulations, suggesting that AI safety is an engineering challenge rather than a legal one. He believes that existing laws and market forces are sufficient to ensure companies release safe AI products. Huang's stance reflects his confidence in the industry's ability to self-regulate and innovate safely without additional legal constraints. However, this perspective may be influenced by Nvidia's vested interest in the AI market, where regulation could potentially slow down growth. The debate continues on whether self-regulation is enough to address AI's potential risks.
TechCrunch AI
© TechCrunch AIMore in Models & Labs
Models & Labscoding
llama.cpp b10996 adds CUDA 13 and ROCm 10.0 builds
This release quietly closes the hardware gap for local inference by adding native support for CUDA 13 and ROCm 10.0 across Linux and Windows. NVIDIA users can now leverage newer driver stacks without waiting for upstream updates, while AMD GPU owners finally get first-class ROCm 10 binaries that match the maturity of their CUDA counterparts. Apple Silicon builds remain available but KleidiAI is explicitly disabled here, suggesting a focus on stability over new kernel optimizations for this specific iteration. The inclusion of openEuler support for Huawei's Ascend chips further broadens the ecosystem beyond standard x86 and ARM consumer hardware.
llama.cpp Releases
Models & Labsmodels
Llama.cpp b10981 release enhances OpenVINO support
The latest b10981 release of llama.cpp brings significant improvements to OpenVINO integration, particularly in optimizing stateful decode and GPU MoE inference. By addressing issues like stateful decode errors and enhancing the handling of sliding-window layers, this update ensures more reliable and efficient model performance. The release also introduces new features such as the GGML_OPENVINO_REQUANT_KQUANT for 4-bit requantization and improved handling of multi-head models. These changes make llama.cpp more robust and versatile, especially for developers working with complex AI models on diverse hardware setups.
