Search
Sort Score
Result 10 results
Languages All
Labels All
Results 41 - 50 of 299 for

gpu

(0.12 sec)

  1. Aspose.LLM for .NET|Documentation

    Aspose.LLM for .NET is a library for on-premise LLM integration, chat sessions, vision input, and model presets....run them locally — on CPU or GPU, without calling a hosted inference...concurrent sessions CPU-only , CUDA GPU , offline , ASP.NET Core Long...

    docs.aspose.com/llm/net/
    more..

  2. Tune for speed vs quality|Documentation

    Move a preset along the speed-quality curve in Aspose.LLM for .NET — model size, quantization, sampler, context, Gpu offload....50-100 tokens/sec on a mid-range GPU, 15-30 on modern CPU. Quality-biased...10-30 tokens/sec on a high-end GPU. CPU not recommended. Balanced...

    docs.aspose.com/llm/net/how-to/tune-for-speed-v...
    more..

  3. UseMemoryMapping|Documentation

    Toggle mmap-based model loading in Aspose.LLM for .NET — default true; disable only for network filesystems or specific memory layouts....copied from the mapped file to GPU memory. What’s next UseMemoryLocking...prevent paging. GpuLayers — GPU offload. Model inference hub...

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

  4. Process of rendering glyphs

    Explore the fundamentals of glyph rendering, covering the processing pipeline, key technologies, and performance considerations....DirectWrite, Core Text, or a custom GPU shader). The rasterizer executes...Compositing, post‑processing & GPU upload Rasterized bitmaps are...

    docs.aspose.com/font/net/what-is-font/glyph-ren...
    more..

  5. API reference|Documentation

    Links to the full class-level API reference and a summary of the key Aspose.LLM for .NET types....ModelInferenceParame Model-load knobs: GPU layers, main GPU, split mode, memory...MultimodalContextPar Vision ( mtmd ) context: GPU use, timings, thread count, verbosity...

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

  6. Low memory tuning|Documentation

    Fit Aspose.LLM for .NET into a tight memory budget — small model, short context, KV cache quantization, aggressive offload, and memory mapping.... Sharing a GPU with other processes, leaving...Llama32Preset (); // CPU-only to avoid GPU memory. preset . BinaryManagerParamet...

    docs.aspose.com/llm/net/use-cases/low-memory-tu...
    more..

  7. Presets|Documentation

    Preset base class, parameter bags, and patterns for using and customizing presets in Aspose.LLM for .NET....ModelInferenceParame Model-load knobs: GPU layers, main GPU, split mode, tensor...MultimodalContextPar mtmd (vision) context: GPU use, timings, thread count, verbosity...

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

  8. Features|Documentation

    Capabilities of Aspose.LLM for .NET — local inference, chat sessions, vision input, acceleration, session persistence, and what the SDK does not do....inference Run models on your CPU or GPU. Input, prompts, and model weights...size, sampler, chat template, GPU layers, quantization) while keeping...

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

  9. Getting started|Documentation

    Install Aspose.LLM for .NET, apply a license, and run your first chat example....restore, CPU-only run, CUDA GPU run. Product overview Developer's...

    docs.aspose.com/llm/net/getting-started/
    more..

  10. Triangulate Mesh|Documentation

    Aspose.3D for Java API has support of triangulating mesh, which is useful for game industry because the triangle is the only supported primitive that Gpu hardware supports(non-triangle data are triangulated in driver-level, which is inefficient in real-time rendering)....only supported primitive that GPU hardware supports(non-triangle...

    docs.aspose.com/3d/java/triangulate-mesh/
    more..