» Tag
nvidia
43 postsBw24: From-Scratch Rust+CUDA LLM Engine Tuned for RTX 50 Blackwell
Bw24 is a from-scratch Rust+CUDA LLM inference engine tuned for RTX 50-series Blackwell GPUs, hitting up to 2.3x speculative decoding gains over llama.cpp.
Clustering Two Dell Pro Max GB10 Systems for Local AI Compute
Two Dell Pro Max GB10 systems clustered via RDMA/RoCE for 256GB of local AI memory, revealing setup quirks and firmware pitfalls.
Inside NVIDIA Exemplar Cloud: Closing 8-12% AI Training Performance Gaps
How NVIDIA diagnosed 8-12% AI training performance gaps on H100 and GB200 NVL72 clusters using perf, Nsight Systems, SMMU tuning, and NUMA fixes.
CommitBrief — AI code reviews, right in your terminal
A provider-agnostic, local-first CLI that reviews your staged changes, a historic range, or a whole GitHub pull request. Zero telemetry, no server. Free and open source.
commitbrief.comNVIDIA's Ising Decoder Slashes Color Code Logical Error Rates 300x+
NVIDIA's open-source Ising Decoder ColorCode 1 Fast cuts color code logical error rates by 347x+ and speeds decoding 7.3x over Chromobius.
NVIDIA Alpamayo 2 Super: 34B Open Vision-Language-Action Model
NVIDIA introduces Alpamayo 2 Super, a 34B open vision-language-action model for robotaxis.
Can Kimi-K3 Model Run on a Single Node of Nvidia B300 X8?
Explore the performance of Kimi-K3 on a single Nvidia B300 x8 node. Insights on architecture and memory management.
How to Self-Host a Validated AI Coding Assistant with NVIDIA NeMo Guardrails
Self-host your AI coding assistant with NVIDIA NeMo Guardrails. Establish a system that prevents errors and ensures traceability.
Chinese modder gets GeForce RTX 4060 working on Huawei Arm workstation
A Chinese modder has managed to run a GeForce RTX 4060 on a Huawei Arm workstation with Windows 11. Learn about the performance challenges faced.
Nvidia Vera CPU: A New Era for AI Data Centers
Nvidia's Vera CPU stands out with its custom design for AI data centers. SPEC CPU 2026 benchmark results showcase its performance.
Benchmarking 15 'E-Waste' GPUs with Modern Workloads
An evaluation of decommissioned NVIDIA Tesla GPUs' performance with modern workloads. Offers crucial insights for low-cost GPU node construction.