Tracked topic
…higher‑precision calculations are generally slower and consume more energy, while narrow-precision formats like NVFP4 are more energy‑efficient and can deliver higher throughput, at equivalent accuracy to FP8.Equally important…
…Related Story NVIDIA and Epic Helped Tides of Annihilation Devs Fix ‘Giant Knight’ Frame Drops Ahead of Summer Demo In terms of specifications, the Rubin CPX features 30 petaFLOPs of NVFP4 compute…
…The precision ladder has a practical destination, and our NVFP4 evaluation helps clarify exactly where each format shines: we tested an NVFP4 path that stores camera features in E2M1 with per-16…
…At sparse NVFP4 precision, the processor is capable of delivering up to 50 PFLOPS of AI compute performance. Memory bandwidth also receives a substantial upgrade through 288 GB of HBM4 memory configured…
Here's a brand new Blackwell-native, prefill-optimized 4-bit quant that runs 50% faster on compatible hardware than a Q4 quant of the same memory footprint. And it runs 4-7% faster than other NVFP4 quants as benchmarked …
These are single stream numbers Following on from my previous post about v100s (here) and inspired by this comment (here) I decided to work on kernels that allow for an extremely fast path for Nvfp4 weights on sm70 and a…
Four Tesla V100s from 2017 matched my RTX 5090 on single-request Qwen 3.8 decode. Repo: https://github.com/dnv2003/v100-skinny https://i.redd.it/5ws2ak3uqckh1.gif The 5090 was not being held back. It ran NInfer, a specia…
Safetensors, llmfan46/Qwen3.5-35B-A3B-uncensored-heretic-v2-Native-MTP-Preserved: https://huggingface.co/llmfan46/Qwen3.5-35B-A3B-uncensored-heretic-v2-Native-MTP-Preserved GGUFs, llmfan46/Qwen3.5-35B-A3B-uncensored-here…
Poolside have updated the FP8 and NVFP4 checkpoints for Laguna S 2.1, increasing the default context size to 1 million, and updating the configs. Here's hoping they fixed the looping issue, this model has been great in m…
…deploying Qwen3-32B with NVFP4 quantization across 64 NVIDIA B200 GPUs, with target SLAs of 1000ms time-to-first-token (TTFT) and 15ms time-per-output-token (TPOT). Using a single command…
…Running both on the same expensive HBM-equipped GPU wastes silicon, so NVIDIA built a dedicated companion: "Rubin CPX." It is a monolithic, single-die accelerator delivering 30 PetaFLOPS of NVFP4 compute…
…Related Story AMD Finally Overtakes Intel in Q1 Data Center Revenue as Agentic AI Forces Hyperscalers to Hoard CPUs Over GPUs Today, NVIDIA has announced that with NVFP4 support, the DGX Spark…
…BF16 및 NVFP4 현재 모델은 BF16 체크포인트와 함께 Hugging Face 에서 제공되며, NVIDIA Model Optimizer 를 사용한 DiffusionGemma 용 NVFP4 양자화 체크포인트도 이용 가능합니다. NVIDIA NIM으로 엔터프라이즈 배포하기 NVIDIA NIM 을 사용하면…
…NVFP4 precision The same NVFP4 checkpoint runs on NVIDIA Hopper, NVIDIA Blackwell, and Ampere GPUs. Developers can use one checkpoint across all NVIDIA GPU architectures thanks to specialized NVFP4 quantization kernels. NVFP4…
…On the other hand, NVIDIA's NVLink, NVFP4, Multi-Token-Prediction, and other technologies also offer meaningful gains, offering a combined 20x throughput increase. NVIDIA’s Blackwell GPUs, powered by continuous full…