Cpython Jit Performance Uplift In Python 3 15 Alpha

Issue 76 Edition 2026-03-17 4 min read

Not accepted General

Sources: 1 • Confidence: Medium • Updated: 2026-04-12 10:16

Key takeaways

In Python 3.15 alpha on macOS AArch64, the JIT is about 11–12% faster than the tail-calling interpreter.
The CPython JIT has already met its stated (modest) performance goals over a year early on macOS AArch64 and a few months early on x86_64 Linux.
In Python 3.15 alpha on x86_64 Linux, the JIT is about 5–6% faster than the standard interpreter.
Python 3.15’s JIT is back on track.

In Python 3.15 alpha on macOS AArch64, the JIT is about 11–12% faster than the tail-calling interpreter.
In Python 3.15 alpha on x86_64 Linux, the JIT is about 5–6% faster than the standard interpreter.

The CPython JIT has already met its stated (modest) performance goals over a year early on macOS AArch64 and a few months early on x86_64 Linux.
Python 3.15’s JIT is back on track.

What benchmark suite(s), configurations, and measurement methodology produced the reported ~11–12% (macOS AArch64) and ~5–6% (x86_64 Linux) speedups in Python 3.15 alpha?
What specifically are the CPython JIT’s stated performance goals, and what criteria determine that they have been met on each platform?
What does “back on track” refer to for the Python 3.15 JIT (missed milestones, regressions, staffing changes, scope changes), and what evidence supports the status change?
Do the reported JIT performance gains persist across subsequent Python 3.15 pre-releases (alpha to beta/RC) and across a broader set of workloads?

Early meeting of modest CPython JIT performance goals on macOS AArch64 and x86_64 Linux could reduce perceived execution risk for the Python 3.15 JIT effort.
If JIT speedups in 3.15 alpha generalize beyond the cited baselines, it could improve Python runtime performance perception, potentially affecting toolchains and workloads sensitive to interpreter speed.

Independent disclosures of the benchmark suites, configurations, and measurement methodology behind the reported 11 to 12 percent and 5 to 6 percent speedups, with reproducible results.
Clear definition of stated JIT performance goals and the criteria for meeting them, reported consistently for each platform.
Subsequent 3.15 pre-releases maintain or improve the reported deltas across a broader workload set, not just the specified interpreter baselines.

Methodology details show results depend on narrow benchmarks, atypical configurations, or incomparable baselines, undermining the reported uplift.
Later 3.15 betas or release candidates show the speedups do not persist or regress materially versus the same baselines.
Clarification reveals performance goals were minimal, redefined, or the back on track status reflects scope changes rather than sustained performance progress.