Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Add an LLM demo to apps/ #8216

Open
wants to merge 20 commits into
base: main
Choose a base branch
from
Open

Add an LLM demo to apps/ #8216

wants to merge 20 commits into from

Commits on Apr 18, 2024

  1. Initial push

    @steven-johnson
    steven-johnson committed Apr 18, 2024
    Configuration menu
    Copy the full SHA
    1fbdf07 View commit details
    Browse the repository at this point in the history

Commits on Apr 24, 2024

  1. Add apps/hallmark. 

    Hallmark (HAlide LLM Advanced Research Kit) is Halide-written execution engine for Gemini/Gemma models; it serves as a testbed for writing efficient ML kernels in Halide. It is incomplete and experimental at this point.
    @steven-johnson
    steven-johnson committed Apr 24, 2024
    Configuration menu
    Copy the full SHA
    c587211 View commit details
    Browse the repository at this point in the history

  2. Merge branch 'srj/apps-hamal' of https://github.com/halide/Halide int… 

    …o srj/apps-hallmark
    @steven-johnson
    steven-johnson committed Apr 24, 2024
    Configuration menu
    Copy the full SHA
    6bed0f5 View commit details
    Browse the repository at this point in the history

Commits on May 1, 2024

  1. Move some files to Halid e.cpp convention rather than .cc .

    Zalman Stern committed May 1, 2024
    Configuration menu
    Copy the full SHA
    16e7619 View commit details
    Browse the repository at this point in the history

Commits on May 3, 2024

  1. Start of command line to generate text from prompt using LLM.

    Zalman Stern committed May 3, 2024
    Configuration menu
    Copy the full SHA
    124a745 View commit details
    Browse the repository at this point in the history

Commits on May 7, 2024

  1. Initial checkin of llm_runner command like to generate text from a pr… 

    …ompt.
    Zalman Stern committed May 7, 2024
    Configuration menu
    Copy the full SHA
    10ad034 View commit details
    Browse the repository at this point in the history

  2. Format

    @steven-johnson
    steven-johnson committed May 7, 2024
    Configuration menu
    Copy the full SHA
    cf1f05e View commit details
    Browse the repository at this point in the history

  3. Error outputs

    @steven-johnson
    steven-johnson committed May 7, 2024
    Configuration menu
    Copy the full SHA
    427773f View commit details
    Browse the repository at this point in the history

Commits on May 8, 2024

  1. Add prompt bracketing. Clean up debug #ifdefs.

    Zalman Stern committed May 8, 2024
    Configuration menu
    Copy the full SHA
    6b79cbe View commit details
    Browse the repository at this point in the history

  2. Merge remote-tracking branch 'refs/remotes/origin/srj/hallmark' into … 

    …srj/hallmark
    Zalman Stern committed May 8, 2024
    Configuration menu
    Copy the full SHA
    5db6136 View commit details
    Browse the repository at this point in the history

Commits on May 9, 2024

  1. Remove mistaken older app addition

    @steven-johnson
    steven-johnson committed May 9, 2024
    Configuration menu
    Copy the full SHA
    47efec2 View commit details
    Browse the repository at this point in the history

  2. clang-format

    @steven-johnson
    steven-johnson committed May 9, 2024
    Configuration menu
    Copy the full SHA
    43d1507 View commit details
    Browse the repository at this point in the history

Commits on May 10, 2024

  1. Avoid ambiguous overload of sqrt in GCC

    @steven-johnson
    steven-johnson committed May 10, 2024
    Configuration menu
    Copy the full SHA
    3b8b3bc View commit details
    Browse the repository at this point in the history

Commits on May 28, 2024

  1. Remove extraneous debugging code.

    Zalman Stern committed May 28, 2024
    Configuration menu
    Copy the full SHA
    07c4598 View commit details
    Browse the repository at this point in the history

  2. Remove parallelization that was causing cached version to fail and do… 

    …esn't seem to help

performance in uncached version.
    Zalman Stern committed May 28, 2024
    Configuration menu
    Copy the full SHA
    5619c7c View commit details
    Browse the repository at this point in the history

  3. Turn on kv caching. Significantly improves performance.

    Zalman Stern committed May 28, 2024
    Configuration menu
    Copy the full SHA
    47ec039 View commit details
    Browse the repository at this point in the history

  4. Formatting fix.

    Zalman Stern committed May 28, 2024
    Configuration menu
    Copy the full SHA
    806d3c6 View commit details
    Browse the repository at this point in the history

  5. Add show_timing flag to runner to show how long various operations 

    take and give a per token time.
    Zalman Stern committed May 28, 2024
    Configuration menu
    Copy the full SHA
    0e99743 View commit details
    Browse the repository at this point in the history

  6. Add support for multiple algorithms, including naive and three pass.

    Zalman Stern committed May 28, 2024
    1 Configuration menu
    Copy the full SHA
    5693e38 View commit details
    Browse the repository at this point in the history

  7. Formatting.

    Zalman Stern committed May 28, 2024
    Configuration menu
    Copy the full SHA
    221f723 View commit details
    Browse the repository at this point in the history