jax-ml/scaling-book

Home for "How To Scale Your Model", a short blog-style textbook about scaling LLMs on TPUs

GitHub repository with 1,072 stars and 162 forks.

Language: HTML

Topics: jax, llm-inference, llms, roofline, tpus

Open provider repository

Latest metric snapshot

2026-06-11: 1,072 stars and 162 forks.

Trending in HTML

  1. 1. eooce/nodejs-argo

    nodejs-argo是一个强大的Argo隧道部署工具，专为PaaS平台和游戏玩具平台设计。它支持多种代理协议（VLESS、VMess、Trojan等），并集成了哪吒探针功能

    GitHub repository with 7,905 stars and 12,649 forks.

    Trending score: 5.12; stars gained: +525; forks gained: +1,047.

    Language: HTML

    Topics: argo, cfip, trojan-ws-tls, tunnel, vless-ws-tls, vmess-ws-tls

  2. 2. tw93/Kami

    👩‍🚒 Good content deserves good paper.

    GitHub repository with 8,587 stars and 410 forks.

    Trending score: 4.98; stars gained: +576; forks gained: +25.

    Language: HTML

  3. 3. shanraisshan/claude-code-best-practice

    from vibe coding to agentic engineering - practice makes claude perfect

    GitHub repository with 57,747 stars and 5,800 forks.

    Trending score: 4.39; stars gained: +150; forks gained: +19.

    Language: HTML

    Topics: agentic-ai, agentic-coding, agentic-engineering, agentic-workflow, ai, ai-agents

  4. 4. alchaincyf/huashu-design

    Huashu Design · HTML-native design skill for Claude Code · Claude Code 里 HTML 原生的设计 skill · 高保真原型 / 幻灯片 / 动画 + 20 设计哲学 + 5 维评审 + MP4 导出 · Agent-agnostic

    GitHub repository with 18,724 stars and 2,333 forks.

    Trending score: 4.09; stars gained: +44; forks gained: +3.

    Language: HTML

  5. 5. nexu-io/html-video

    Programmatic video for coding agents — HTML to video on your laptop. Turn HTML, CSS & data into real MP4s with pluggable render engines, 21 templates, AI soundtrack. Apache-2.0, no per-render fees. An official project by the Open Design team.

    GitHub repository with 3,131 stars and 374 forks.

    Trending score: 4.07; stars gained: +69; forks gained: +11.

    Language: HTML

    Topics: ai-agent, apache-2, coding-agent, css, ffmpeg, html

  6. 6. lijigang/ljg-skills

    GitHub repository with 5,818 stars and 674 forks.

    Trending score: 3.96; stars gained: +113; forks gained: +15.

    Language: HTML

Trending topic: jax

  1. 1. jax-ml/jax

    Composable transformations of Python+NumPy programs: differentiate, vectorize, JIT to GPU/TPU, and more

    GitHub repository with 35,821 stars and 3,636 forks.

    Trending score: 2.44; stars gained: +10; forks gained: +6.

    Language: Python

    Topics: jax

  2. 2. zengxiao-he/tessera

    From teacher to tiles — a from-scratch LLM distillation & serving engine: custom Triton/CUDA kernels, FSDP distillation, paged-KV continuous batching, speculative decoding, a Rust gateway, a JAX oracle, and interpretability tooling.

    GitHub repository with 179 stars and 1 forks.

    Trending score: 2.19; stars gained: +11; forks gained: +0.

    Language: Python

    Topics: cuda, flash-attention, fsdp, inference-engine, jax, knowledge-distillation

  3. 3. AI-Hypercomputer/maxtext

    A simple, performant and scalable Jax LLM!

    GitHub repository with 2,322 stars and 534 forks.

    Trending score: 1.82; stars gained: +7; forks gained: +3.

    Language: Python

    Topics: deepseek, fine-tuning, gemma2, gemma3, gpt, jax

  4. 4. pnkraemer/probdiffeq

    Probabilistic solvers for differential equations in JAX. Adaptive ODE solvers with calibration, state-space model factorisations, and custom information operators. Compatible with the broader JAX scientific computing ecosystem.

    GitHub repository with 61 stars and 3 forks.

    Trending score: 0.63; stars gained: +1; forks gained: +0.

    Language: Python

    Topics: differential-equations-solvers, jax, scientific-computing, probabilistic-numerics

  5. 5. google/orbax

    Orbax provides common checkpointing and persistence utilities for JAX users

    GitHub repository with 518 stars and 99 forks.

    Trending score: 0.50; stars gained: +0; forks gained: +1.

    Language: Python

    Topics: checkpoint, flax, jax

  6. 6. AshiteshSingh/Tpu-Accelerated-Quantum-JAX

    High-performance, differentiable quantum state-vector & tensor network simulator in 100% pure JAX (no classical framework overhead). Accelerated on NVIDIA GPUs and Google Cloud TPU v6e-64/v5e VM clusters up to 37 qubits! Supported by Google's TPU Research Cloud (TRC) program.

    GitHub repository with 25 stars and 2 forks.

    Trending score: 0.47; stars gained: -1; forks gained: +0.

    Language: Jupyter Notebook

    Topics: google-cloud, grovers-algorithm, jax, machine-learning, nvidia-gpu, python