約82件1ページ目

日本語のみで絞り込む

条件を指定して検索しています。すべての条件を解除する

  • 最終更新日:24時間以内
  • 22時間前 -GPU benchmark software tutorial: run 3DMark, Superposition and llama.cpp v1.0.0, log nvidia-smi telemetry and compare runs with a Python harness.

    20時間前 -Build a repeatable NVIDIA GPU benchmark workflow for Windows 11 or Linux using graphics tests, llama-bench, and telemetry while changing one variable at a ...

    6時間前 -Run MoE models bigger than your RAM. Frontier-size MoE on a phone, laptop, edge. Lossless, on stock llama.cpp. An adaptive streaming optimizer.

    10時間前 -The motivating consumer is llama.cpp/ggml, where building a cuDNN graph cost about 30 ms per new shape. Experimental: the C ABI may still change before the ...

    17時間前 -With the vision projector and a 32K context, llama-server held 51.4 GB of GPU memory on our CUDA box. We have not measured it on a Mac. Quality. Measured ...

    11時間前 -I made a Windows GUI to manage, benchmark and compare multiple llama.cpp builds — handy for AMD GPU users. 32. 10. Finally got PyTorch running smoothly on AMD ...

    21時間前 -Eager mode and CUDA graphs. In eager mode, the engine launches each GPU kernel of a decode step from Python, so many small kernels cost CPU time. With CUDA ...

    14時間前 -RTX Spark runs Windows on Arm. PyTorch provides a CUDA install path for it, but that does not mean every Linux ARM64 training package, FlashAttention build, ...

    21時間前 -Merge a LoRA adapter into its base model, export GGUF with llama.cpp v1.0.0, quantize to Q4_K_M and run it in Ollama. 12 steps, code and fixes.

    21時間前 -An all-in-one, pure C++ inference engine for audio models, powered by ggml. Supports TTS, STT, VAD, voice conversion, music generation, and more, with highly ...