Quoting Ken Jin

2026-03-17 · Source: Simon Willison's Weblog · Field: Technology & Digital — Software Development & Engineering, Artificial Intelligence & Machine Learning · Depth: Intermediate, quick

Summary

The CPython JIT project has successfully met its performance goals ahead of schedule for both macOS AArch64 and x86_64 Linux, as reported on March 17, 2026. The Python 3.15 alpha JIT now shows an 11-12% speed increase on macOS AArch64 over the tail calling interpreter. Additionally, it is 5-6% faster than the standard interpreter on x86_64 Linux, marking significant early progress in Python's runtime optimization.

Key takeaway

The Python 3.15 alpha JIT compiler now delivers 11-12% faster execution on macOS AArch64 and 5-6% faster on x86_64 Linux. This early success in CPython's JIT development signals a substantial performance boost for data processing and AI/ML workloads reliant on native Python execution.

Topics

• CPython JIT
• Python 3.15
• Performance Optimization
• macOS AArch64
• x86_64 Linux

Best for: Machine Learning Engineer, NLP Engineer, Computer Vision Engineer, Software Engineer, Research Scientist, AI Engineer

Related on AIssential

• [P] Bypassing CoreML to natively train a 110M Transformer on the Apple Neural Engine (Orion) — ORION is introduced as the first open end-to-end system enabling direct Apple Neural Engine (ANE) execution and stable multi-step training for large language models, bypassing CoreML's limitations.
• Topping the GPU MODE Kernel Leaderboard with NVIDIA cuda.compute — The `cuda.compute` library enables Python developers to achieve near "speed-of-light" GPU performance with CUB primitives.
• Vibe coding is about to get so fast — GPT 5.3 Spark leverages Cerebras wafer-scale chips for unprecedented coding inference speed.
• TorchTPU: Running PyTorch Natively on TPUs at Google Scale — TorchTPU enables native PyTorch execution on Google TPUs, prioritizing usability, performance, and hardware portability.
• From Naive to Near-Peak: Building High-Performance GEMM Kernels with Gluon — Achieving near-peak GPU kernel performance requires explicit hardware control and a profiling-driven, iterative optimization methodology.
• I Compared 4 Python HTTP Libraries. One Shocked Me Completely. — The author, after experiencing significant performance bottlenecks with the `requests` library in a high-throughput system making hundreds of API calls per minute, undertook a comprehensive benchmark…

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Simon Willison's Weblog.