CPython JIT Development Paused Pending Standards Track PEP

The Python Steering Council has formally requested a pause on new development for the experimental just-in-time (JIT) compiler in CPython. New features, optimizations, and performance work may not land on the main branch until a formal Standards Track PEP (Python Enhancement Proposal) is authored and accepted. Bugfixes and security fixes will continue as normal.

Formalization of JIT Status

The Steering Council's decision is driven by the necessity to move the JIT project from an experimental status to a supported, non-experimental part of CPython. While PEP 744 provided an initial design, it was an Informational PEP, and several critical open questions regarding long-term maintenance, security reviews, and tooling support remained unresolved.

To transition the JIT to a permanent feature, the Council requires a new Standards Track PEP that addresses the following key areas:

  • Long-term Maintenance: A clear plan for how the JIT subsystem will be sustained without placing undue burden on maintainers who do not directly contribute to the JIT.
  • Compatibility and Tooling: Guarantees on how the JIT interacts with existing CPython features, including free-threading, profilers, and debuggers.
  • Success Metrics: Defined, measurable targets for performance, platform coverage, and memory overhead, along with associated timelines.
  • Ecosystem Integration: The relationship between the official JIT and third-party implementations such as Numba, PyTorch, and CinderX.
  • Architectural Stability: A determination of whether the current JIT architecture is considered stable or likely to undergo further changes.

Implementation Timeline and Consequences

The Steering Council has established a six-month window for a PEP to be submitted and resolved. If no such PEP is accepted within this timeframe, the JIT code must be removed from the main branch of CPython and any further development must occur outside the main repository.

Community and Developer Perspectives

Community reaction to the announcement has been polarized. Some developers view the move as a necessary application of engineering rigor and governance to a complex subsystem.

"CPython's selling point was that it is simple, fast enough with C extensions and the code was accessible. Complicating the code base for occasional 50% speedups... just isn't worth it."

Conversely, other contributors and observers argue that the pause in development momentum could jeopardize the project's success.

"Losing development momentum for a beancounting reason like this one is a sure way to kill a project."

Some observers have also suggested that the Council's openness to alternative JIT implementation strategies—or a more pluggable JIT infrastructure—might indicate a lack of confidence in the current specific implementation, potentially leading to its removal if a more flexible architecture is proposed.

Sources