Install
Frameworks & Training
PyTorch, TensorFlow, JAX, datasets, and accelerator workflows.
- 6 Tracked terms
- Last 30 days Feed window
What this topic collects on
An article joins this feed when it matches these terms. Each one is also a search of its own.
Related topics
Latest in Frameworks & Training
Serverless GPUs for up to 79% less than Modal.
8+ hour, 45+ min ago (618+ words) Run your Python or your container on GPUs in modular data centres we design and build ourselves. Per-second billing, and reserved capacity at a published rate from $0.63 per GPU-hour. Every GPU both platforms sell, list price to list price. On…...
How DOCA GPUNetIO Unifies GPU-Initiated Networking Across the NVIDIA Software Stack
6+ hour, 19+ min ago (1054+ words) GPU applications increasingly need networking and data movement to behave like first-class GPU-controlled operations rather than host-driven services. When the CPU sits in the middle of every network transaction, it becomes a bottleneck on the critical path, adding latency and…...
AICR v1.0: Open, stable, and verifiable GPU cluster configuration
9+ hour, 13+ min ago (494+ words) GPU-accelerated Kubernetes clusters depend on compatible versions across dozens of components, each on its own release cycle: host kernels, GPU drivers, container runtimes, networking, storage, operators, and workload frameworks. A configuration that works for one service, GPU generation, and Kubernetes…...
Control How Your GPU Shares Work with Green Contexts
1+ day, 3+ hour ago (565+ words) GPU applications increasingly consist of multiple independent components running at the same time within a single process: a latency-sensitive operator alongside a throughput-oriented background kernel; a data preprocessing stage alongside model inference; or multiple stages of a processing workflow sharing…...
Someone Trained a Transformer Without Backpropagation. I Rebuilt the Core Trick in Under 60 Lines of NumPy.
20+ hour, 53+ min ago (222+ words) Dust trains a language model using only forward passes and noise. It’s not faster than … Someone Trained a Transformer Without Backpropagation. I Rebuilt the Core Trick in Under 60 Lines of NumPy. Dust trains a language model using only forward passes…...
C3 AI Accelerator Program End User License Agreement
1+ day, 3+ hour ago (655+ words) THIS C3 AI END USER LICENSE AGREEMENT (“Agreement”) governs Your use of the C3 AI Materials in connection with the C3 AI Accelerator Program. C3.ai, Inc. is referred to as “C3 AI,” “We,” “Us” and “Our.” 1.2 “C3 AI Documentation” means the user documentation made available…...
Auto-cpufreq 3.2 Adds Intel HWP Dynamic Boost, Lands in Debian Repos
1+ day, 8+ hour ago (334+ words) The Linux CPU power optimization tool auto-cpufreq 3.2 brings new Intel HWP features, improved battery reporting, and Debian repository support. Linux CPU power optimization tool auto-cpufreq has released version 3.2 and is now available as a.deb package in Debian’s official testing repositories,…...
A 125B Model on a 12 GB GPU? Here’s Where Qwen 3.8 Flash Next’s 24,576 Experts Actually Live
1+ day, 2+ hour ago (250+ words) Strata runs Alibaba’s open-weight MoE on a gaming PC. Before you download 66 GB, run the same memory … A 125B Model on a 12 GB GPU? Here’s Where Qwen 3.8 Flash Next’s 24,576 Experts Actually Live Strata runs Alibaba’s open-weight MoE on a gaming PC. Before…...
Running a 125B-Parameter LLM on a Home PC: "Strata" Opens a New Frontier for Local AI
1+ day, 2+ hour ago (312+ words) Strata, released by developer Niko1221, adopts a design that loads the entire MoE (Mixture of Experts) model into system RAM while placing only the most frequently used expert sub-models into GPU VRAM. By combining this with speculative decoding—where a lightweight…...
Clockwork.io Raises $31M as LinkedIn, Together AI and WhiteFiber Adopt Its Resilience Software to Stop Wasting GPU-Hours
1+ day, 12+ hour ago (442+ words) Oct 05, 2026, 09:00 ET LinkedIn prevents tens of thousands of GPU-hours of downtime monthly; new TorchPass innovations preserve AI workload progress without code changes and speed reinforcement learning. PALO ALTO, Calif., Oct. 5, 2026 /PRNewswire/ -- Clockwork.io, whose fault-tolerance software keeps AI training, reinforcement…...