bm-20240605-3.13.0b2-3a83b17

Results

• fork: python
• version: 3.13.0b2
• config:
• commit hash: 3a83b17
• commit date: 2024-06-05T16:46:34+02:00
• ref: v3.13.0b2

linux aarch64 (arminc)

• GitHub Action run
• cpu model: missing
• platform: Linux-5.15.0-101-generic-aarch64-with-glibc2.35
• raw results

vs. 3.10.4

• Geometric mean: 1.28x faster (HPT: reliability of 100.00%, 1.22x faster at 99th %ile)
• Memory usage: 1.13x
• missing benchmarks: sqlalchemy_declarative, sqlalchemy_imperative
• new benchmarks: async_tree_cpu_io_mixed_tg, async_tree_io_tg, async_tree_memoization_tg, async_tree_none_tg, bpe_tokeniser
• 📄table
• 📈time plot

vs. 3.11.0

• Geometric mean: 1.05x faster (HPT: reliability of 99.80%, 1.00x faster at 99th %ile)
• Memory usage: 1.04x
• missing benchmarks: sqlalchemy_declarative, sqlalchemy_imperative
• new benchmarks: bpe_tokeniser
• 📄table
• 📈time plot

vs. 3.12.0

• Geometric mean: 1.01x faster (HPT: reliability of 83.98%, 1.00x faster at 99th %ile)
• Memory usage: 0.92x
• missing benchmarks: sqlalchemy_declarative, sqlalchemy_imperative
• new benchmarks: bpe_tokeniser, flaskblogging
• 📄table
• 📈time plot

linux x86_64 (linux)

• GitHub Action run
• cpu model: Intel(R) Xeon(R) W-2255 CPU @ 3.70GHz
• platform: Linux-5.4.0-164-generic-x86_64-with-glibc2.31
• raw results

vs. 3.10.4

• Geometric mean: 1.32x faster (HPT: reliability of 100.00%, 1.24x faster at 99th %ile)
• Memory usage: 1.12x
• missing benchmarks: sqlalchemy_declarative, sqlalchemy_imperative, unpack_sequence
• new benchmarks: async_tree_cpu_io_mixed_tg, async_tree_io_tg, async_tree_memoization_tg, async_tree_none_tg, bpe_tokeniser
• 📄table
• 📈time plot

vs. 3.11.0

• Geometric mean: 1.05x faster (HPT: reliability of 84.61%, 1.00x faster at 99th %ile)
• Memory usage: 1.04x
• missing benchmarks: sqlalchemy_declarative, sqlalchemy_imperative, unpack_sequence
• new benchmarks: bpe_tokeniser
• 📄table
• 📈time plot

vs. 3.12.0

• Geometric mean: 1.03x faster (HPT: reliability of 98.90%, 1.00x faster at 99th %ile)
• Memory usage: 0.98x
• missing benchmarks: sqlalchemy_declarative, sqlalchemy_imperative, unpack_sequence
• new benchmarks: bpe_tokeniser, djangocms, flaskblogging, genshi_text, genshi_xml, html5lib, pylint, thrift
• 📄table
• 📈time plot

linux x86_64 (pythonperf2)

• GitHub Action run
• cpu model: 12th Gen Intel(R) Core(TM) i9-12900
• platform: Linux-5.15.0-105-generic-x86_64-with-glibc2.35
• raw results

vs. 3.10.4

• Geometric mean: 1.28x faster (HPT: reliability of 100.00%, 1.21x faster at 99th %ile)
• Memory usage: 1.13x
• missing benchmarks: sqlalchemy_declarative, sqlalchemy_imperative, unpack_sequence
• new benchmarks: async_tree_cpu_io_mixed_tg, async_tree_io_tg, async_tree_memoization_tg, async_tree_none_tg, bpe_tokeniser
• 📄table
• 📈time plot

vs. 3.11.0

• Geometric mean: 1.07x faster (HPT: reliability of 99.31%, 1.00x faster at 99th %ile)
• Memory usage: 1.04x
• missing benchmarks: sqlalchemy_declarative, sqlalchemy_imperative, unpack_sequence
• new benchmarks: bpe_tokeniser
• 📄table
• 📈time plot

vs. 3.12.0

• Geometric mean: 1.00x slower (HPT: reliability of 70.35%, 1.00x slower at 99th %ile)
• Memory usage: 0.93x
• missing benchmarks: sqlalchemy_declarative, sqlalchemy_imperative, unpack_sequence
• new benchmarks: bpe_tokeniser, flaskblogging, genshi_text, genshi_xml, html5lib, pylint, thrift
• 📄table
• 📈time plot

windows amd64 (pythonperf1)

• GitHub Action run
• cpu model: missing
• platform: Windows-11-10.0.22631-SP0
• raw results

vs. 3.10.4

• Geometric mean: 1.23x faster (HPT: reliability of 100.00%, 1.19x faster at 99th %ile)
• Memory usage: unknown
• missing benchmarks: dask, sqlalchemy_declarative, sqlalchemy_imperative, unpack_sequence
• new benchmarks: async_tree_cpu_io_mixed_tg, async_tree_io_tg, async_tree_memoization_tg, async_tree_none_tg
• 📄table
• 📈time plot

vs. 3.11.0

• Geometric mean: 1.11x faster (HPT: reliability of 100.00%, 1.08x faster at 99th %ile)
• Memory usage: unknown
• missing benchmarks: dask, sqlalchemy_declarative, sqlalchemy_imperative, unpack_sequence
• 📄table
• 📈time plot

vs. 3.12.0

• Geometric mean: 1.08x faster (HPT: reliability of 100.00%, 1.05x faster at 99th %ile)
• Memory usage: unknown
• missing benchmarks: dask, sqlalchemy_declarative, sqlalchemy_imperative, unpack_sequence
• new benchmarks: flaskblogging, genshi_text, genshi_xml, html5lib, pylint, thrift
• 📄table
• 📈time plot

windows x86 (pythonperf1_win32)

• GitHub Action run
• cpu model: missing
• platform: Windows-11-10.0.22631-SP0
• raw results

vs. 3.10.4

• Geometric mean: 1.15x faster (HPT: reliability of 100.00%, 1.11x faster at 99th %ile)
• Memory usage: unknown
• missing benchmarks: aiohttp, dask, dulwich_log, mypy2, sqlalchemy_declarative, sqlalchemy_imperative, unpack_sequence
• new benchmarks: async_tree_cpu_io_mixed_tg, async_tree_io_tg, async_tree_memoization_tg, async_tree_none_tg
• 📄table
• 📈time plot

vs. 3.11.0

• Geometric mean: 1.26x faster (HPT: reliability of 100.00%, 1.25x faster at 99th %ile)
• Memory usage: unknown
• missing benchmarks: aiohttp, dask, dulwich_log, mypy2, sqlalchemy_declarative, sqlalchemy_imperative, unpack_sequence
• 📄table
• 📈time plot

vs. 3.12.0

• Geometric mean: 1.19x faster (HPT: reliability of 100.00%, 1.18x faster at 99th %ile)
• Memory usage: unknown
• missing benchmarks: aiohttp, dask, dulwich_log, mypy2, sqlalchemy_declarative, sqlalchemy_imperative, unpack_sequence
• new benchmarks: flaskblogging, genshi_text, genshi_xml, html5lib, pylint, thrift
• 📄table
• 📈time plot

darwin arm64 (darwin)

• GitHub Action run
• cpu model: missing
• platform: macOS-14.5-arm64-arm-64bit-Mach-O
• raw results

vs. 3.10.4

• Geometric mean: 1.27x faster (HPT: reliability of 100.00%, 1.21x faster at 99th %ile)
• Memory usage: 1.18x
• missing benchmarks: sqlalchemy_declarative, sqlalchemy_imperative, unpack_sequence
• new benchmarks: async_tree_cpu_io_mixed_tg, async_tree_eager, async_tree_eager_cpu_io_mixed, async_tree_eager_cpu_io_mixed_tg, async_tree_eager_io, async_tree_eager_io_tg, async_tree_eager_memoization, async_tree_eager_memoization_tg, async_tree_eager_tg, async_tree_io_tg, async_tree_memoization_tg, async_tree_none_tg, bpe_tokeniser
• 📄table
• 📈time plot

vs. 3.11.0

• Geometric mean: 1.04x faster (HPT: reliability of 93.86%, 1.00x faster at 99th %ile)
• Memory usage: 1.11x
• missing benchmarks: sqlalchemy_declarative, sqlalchemy_imperative, unpack_sequence
• new benchmarks: async_tree_eager, async_tree_eager_cpu_io_mixed, async_tree_eager_cpu_io_mixed_tg, async_tree_eager_io, async_tree_eager_io_tg, async_tree_eager_memoization, async_tree_eager_memoization_tg, async_tree_eager_tg, bpe_tokeniser
• 📄table
• 📈time plot

vs. 3.12.0

• Geometric mean: 1.08x faster (HPT: reliability of 100.00%, 1.04x faster at 99th %ile)
• Memory usage: 1.07x
• missing benchmarks: sqlalchemy_declarative, sqlalchemy_imperative, unpack_sequence
• new benchmarks: async_tree_eager, async_tree_eager_cpu_io_mixed, async_tree_eager_cpu_io_mixed_tg, async_tree_eager_io, async_tree_eager_io_tg, async_tree_eager_memoization, async_tree_eager_memoization_tg, async_tree_eager_tg, bpe_tokeniser, flaskblogging, genshi_text, genshi_xml, html5lib, pylint, thrift
• 📄table
• 📈time plot