NVIDIA released the Nemotron-H family of hybrid Mamba-Transformer models, including 8B models (Nemotron-H-8B-Base, Nemotron-H-8B-Instruct, Nemotron-H-8B-VLM) and 47-56B models (Nemotron-H-47B-Base, Ne…
A user reports achieving 407 FPS in Batman: Arkham Knight at 4K with max visuals and PhysX by using a dual-GPU setup with two RTX 5090s, running DLSS 5 on the second GPU via a Neural CoProcessor, and …
A developer trained a small transformer from scratch in 1.5 hours on an NVIDIA RTX 5090 for 67 cents of compute, scoring 44% on the ARC-AGI-1 benchmark and 7% on ARC-2, matching the performance of TRM…
A developer running a local LLM environment with an RTX 5090 32GB, AMD Strix Halo 128GB, and RTX 6000 Pro Blackwell 96GB reports that CPU and RAM are nearly irrelevant for fully offloaded workloads, w…
Autonomous Computer 2, a 2-GPU AI workstation from Autonomous Computer, is priced at $25,317 for 3-4 units, with volume discounts down to $24,273 for 30+ units, and features dual RTX 5090 or RTX PRO 6…
In a head-to-head test, Anthropic's Claude Opus 5 and Alibaba's Qwen3.8 27B (running locally on an NVIDIA RTX 5090) analyzed the validity of Apple's US Patent 7,844,915 (the 'pinch-to-zoom' patent) ag…
A user running Qwen3.8-27B on an NVIDIA RTX 5090 32 GB with LM Studio reports decode speeds ranging from 12.38 to 47.4 tok/s depending on quantization and multi-token prediction settings, with the Q4_…
SGLang, the open-source inference project associated with Ying Sheng and Banghua Zhu, published deployment recipes for Alibaba's Qwen3.8-27B model, enabling single-GPU serving with NVFP4 quantization …
Meta released Muse Glimmer, a 30-billion-parameter open-weight model for local agentic work, with a 4-bit version fitting under 20 GB and tested within 24 GB or 32 GB memory envelopes on consumer hard…
Newegg is offering a $500 discount on a Skytech Gaming O11 Vision prebuilt PC featuring an NVIDIA RTX 5090 32GB GPU and an AMD Ryzen 7 9800X3D processor, reducing the price to an unspecified amount fr…
RunInfra enables running Kimi-Linear-48B, a distilled version of the full 2.78-trillion-parameter Kimi K3 model, on a single consumer GPU such as the RTX 5090 with 32 GB VRAM, achieving 113.83 tokens …
Jeff Cui and collaborators announced Patch Policy, a robot-control method that beats OpenVLA-OFT with 0.7% of its parameters and trains on a single NVIDIA RTX 5090 GPU. The approach freezes a Vision T…
Thinking Machines' open Inkling model, with 975 billion total parameters and 41 billion active, fits on a single high-memory box at 2-bit or 3-bit quantization, unlike the datacenter-scale Kimi K3. Re…
Unified memory in mini PCs like the AMD Ryzen AI Max+ 395 allows them to run 70-billion-parameter models that exceed the VRAM capacity of high-end GPUs like the NVIDIA RTX 5090, but at significantly s…
A developer migrated a Ryzen 9 9950X3D and RTX 5090 from a full tower to an SFF open-frame custom loop with dual 240mm radiators. The system, named AI-NT-No-Problem, runs local AI inference, remote de…
FastVideo released FastWan-QAD, a family of video generation models that can produce a 5-second 480P video in 1.78 seconds on a single NVIDIA RTX 5090 using quantization-aware distillation. The models…