Skip to content
@ScalingIntelligence

Scaling Intelligence Lab

AI and Systems Laboratory led by Professor Azalia Mirhoseini

Pinned Loading

  1. large_language_monkeys large_language_monkeys Public

    Python 112 26

  2. Archon Archon Public

    Archon provides a modular framework for combining different inference-time techniques and LMs with just a JSON config file.

    Python 190 21

  3. KernelBench KernelBench Public

    KernelBench: Can LLMs Write GPU Kernels? - Benchmark + Toolkit with Torch -> CUDA (+ more DSLs)

    Jupyter Notebook 739 106

  4. tokasaurus tokasaurus Public

    Python 460 34

Repositories

Showing 10 of 15 repositories
  • KernelBench Public

    KernelBench: Can LLMs Write GPU Kernels? - Benchmark + Toolkit with Torch -> CUDA (+ more DSLs)

    ScalingIntelligence/KernelBench’s past year of commit activity
    Jupyter Notebook 739 106 16 (2 issues need help) 11 Updated Jan 7, 2026
  • tokasaurus Public
    ScalingIntelligence/tokasaurus’s past year of commit activity
    Python 460 Apache-2.0 34 3 1 Updated Nov 25, 2025
  • forge-grpo-crusoe Public Forked from allenwang28/forge

    PyTorch-native post-training at scale

    ScalingIntelligence/forge-grpo-crusoe’s past year of commit activity
    Python 0 BSD-3-Clause 72 0 0 Updated Nov 22, 2025
  • ScalingIntelligence/scalingintelligence.github.io’s past year of commit activity
    SCSS 3 19 0 0 Updated Nov 13, 2025
  • good-kernels Public

    Samples of good AI generated CUDA kernels

    ScalingIntelligence/good-kernels’s past year of commit activity
    Python 99 10 1 0 Updated May 30, 2025
  • TPT Public

    Welcome to TPT, a framework for teaching large language models to solve math problems by learning from (and improving on) their own reasoning traces.

    ScalingIntelligence/TPT’s past year of commit activity
    Python 8 4 0 0 Updated May 29, 2025
  • caesar Public

    Throughput-oriented multi-turn inference engine for KernelBench [ICML '25]

    ScalingIntelligence/caesar’s past year of commit activity
    Python 20 8 1 0 Updated May 27, 2025
  • Archon Public

    Archon provides a modular framework for combining different inference-time techniques and LMs with just a JSON config file.

    ScalingIntelligence/Archon’s past year of commit activity
    Python 190 Apache-2.0 21 3 0 Updated Mar 7, 2025
  • codemonkeys Public
    ScalingIntelligence/codemonkeys’s past year of commit activity
    Python 59 MIT 2 2 0 Updated Jan 28, 2025
  • ScalingIntelligence/KernelBenchLeaderboard’s past year of commit activity
    0 0 0 0 Updated Dec 3, 2024

Most used topics

Loading…