Search
Sort Score
Result 10 results
Languages All
Labels All
Results 61 - 70 of 304 for

gpu

(0.18 sec)

  1. Architecture|Documentation

    Understand the layers, runtime flow, and lifecycle of Aspose.LLM for .NET, from the facade API down to native llama.cpp binaries....depending on disk speed and whether GPU offload is active. Memory footprint...sampler state, tokenizer. Both GPU and CPU memory are claimed in...

    docs.aspose.com/llm/net/product-overview/archit...
    more..

  2. 3D 几何图形的基于硬件的呈现|Documentation

    使用 Aspose.3D for .NET API，开发人员可以对Gpu (图形处理单元) 进行编程，并设置图形硬件来渲染 3D 几何图形，而不是固定函数管道。...NET API，开发人员可以对GPU (图形处理单元) 进行编程，并设置图形硬件来渲染 3D ...-它定义了如何将三角形从世界空间转换为屏幕空间并计算GPU侧的最终像素颜色 渲染状态 -它为GPU提供了参数，以将三角形光栅化为像素。 我们准备了一个演示项目。请参阅...

    docs.aspose.com/3d/zh/net/hardware-based-render...
    more..

  3. Context parameters|Documentation

    Reference for Aspose.LLM for .NET context parameters — context size, batch, threading, rope/YaRN scaling, flash attention, KV cache types....KV cache itself live on the GPU. When false , they stay on CPU...varies by build; usually true on GPU-enabled builds. Disable only...

    docs.aspose.com/llm/net/developer-reference/par...
    more..

  4. Installing advanced OCR features|Documentation

    How to extend Aspose.OCR for .NET functionality by adding downloadable modules to the project.... Installing GPU-accelearated package Analyzing...

    docs.aspose.com/ocr/net/modules/
    more..

  5. Streaming-like responses|Documentation

    Aspose.LLM for .NET does not expose token streaming. This page documents the limitation, the workarounds, and the patterns that deliver a responsive user experience without streaming.... On a modern GPU: Short replies (64 tokens) —...completes in under a second on GPU, which is within the threshold...

    docs.aspose.com/llm/net/use-cases/streaming-lik...
    more..

  6. Installing advanced OCR features|Documentation

    How to extend Aspose.OCR for Java functionality by adding downloadable modules to the project.... Installing GPU-accelearated package Analyzing...

    docs.aspose.com/ocr/java/modules/
    more..

  7. CPU|Documentation

    Run Aspose.LLM for .NET on CPU only — AVX512, AVX2, and NoAVX variants, threading, and performance expectations....NET when no supported GPU is available. It works on every...runtimes. Throughput is lower than GPU inference, but reasonable for...

    docs.aspose.com/llm/net/developer-reference/acc...
    more..

  8. AsposeAI class|Documentation

    Reference guide for the AsposeAI class for enabling spell-check correction using LLM models.... gpu_layers int Number of GPU layers to use for...

    docs.aspose.com/ocr/python-net/ai/
    more..

  9. NThreads|Documentation

    CPU thread count for per-token generation in Aspose.LLM for .NET — affects throughput during sequential decoding....NThreads has no effect when GPU offload is active for every layer...

    docs.aspose.com/llm/net/developer-reference/par...
    more..

  10. Batch processing|Documentation

    Run many prompts through one loaded Aspose.LLM model, amortize the load cost, pick session-per-prompt vs shared-session patterns.... GPU offload : batch jobs benefit the most from GPU; a 20×...

    docs.aspose.com/llm/net/use-cases/batch-process...
    more..