XTuner

license

Apache-2.0(OSI)

source

public(InternLM/xtuner)

core-gated

ungated(no enterprise or ee path in the repository root and no paid tier described in the README)

Apache-2.0, an OSI licence, with the full source public and the shipped code in this repository. The LICENSE body was read in full: the stock Apache-2.0 text, its core byte-identical to the canonical template, with nothing after END OF TERMS AND CONDITIONS but the standard appendix. No enterprise or ee path in the repository root and no paid tier described in the readme.

• https://raw.githubusercontent.com/InternLM/xtuner/main/LICENSE recorded 2026-08-31

  LICENSE body, read in full: the stock Apache License 2.0 text, carrying no appended term or additional condition.

• https://raw.githubusercontent.com/InternLM/xtuner/main/README.md recorded 2026-08-31

  README, read for the distribution and gating question: it describes installing and running the code from this repository and names no enterprise, paid or licence-gated build.

What changed here is the instrument, not the level. The PyPI package publishes 0.2.0, uploaded 2025-07-11 and 416 days old, while the repository has since released v1.0.1 on 2026-05-15 and the README documents XTuner V1. The registry never followed the project onto that line, so a count off 0.2.0 measures the version the documentation has moved past. GitHub labels v1.0.1 a pre-release even though the version string reads stable, which is the publisher's own checkbox rather than a claim about the version; the documented current release is V1 either way. A usage_volume record claims to be a download count of the product, and this one counted a superseded version, which is the wrong claim rather than a wrong number. The record therefore states no count: reported_traction, no numeric reach, and the 1,159 downloads kept only as corroboration of the trailing channel. The level is deliberately left where the earlier reading put it. Raising it is a separate judgment needing its own standing evidence, and the only other signal available is the star count, which the routing precedence exists to keep out of an adoption band.

• https://pypi.org/pypi/xtuner/json recorded 2026-08-31

  info.version 0.2.0, uploaded 2025-07-11T05:43:58Z, newest of 27 releases; nothing on a 1.x line has ever been published to the registry.

• https://api.github.com/repos/InternLM/xtuner/releases?per_page=3 recorded 2026-08-31

  Newest three releases: v1.0.1 published 2026-05-15T06:57:47Z (prerelease true, draft false), v1.0.0rc0 of 2025-11-18, and v0.2.0 of 2025-07-11T05:43:37Z - the last release the registry matched, published the same day as the PyPI upload.

• https://pypistats.org/api/packages/xtuner/recent recorded 2026-08-31

  last_month 1159, last_week 112, last_day 11 for the xtuner package. Recorded as a measurement of the trailing registry line, explicitly NOT as the basis of the band.

• https://api.github.com/repos/InternLM/xtuner recorded 2026-08-31

  Repo metadata for InternLM/xtuner: stargazers_count 5187, archived false, pushed_at 2026-08-31 - the attention figure the note contrasts with the download count.

XTuner V1's key-features section claims frontier MoE scale - 1T parameters, 200B without expert parallelism, 64k sequences without sequence parallelism - which is the scale story that places megatron-lm at 5. It sits one band below because Megatron-LM offers composable DP/TP/PP/CP/EP with Megatron-FSDP across model families and publishes weak- and strong-scaling benchmarks, whereas XTuner is specialized to MoE and its throughput claim is stated in the README rather than in a published scaling study. The same one_below relation places swift and llama-factory.

• https://raw.githubusercontent.com/InternLM/xtuner/main/README.md recorded 2026-08-31

  Key Features section states dropless training of 200B MoE without expert parallelism and 600B with intra-node EP, 64k-sequence training on 200B MoE without sequence parallelism, DeepSpeed Ulysses support, MoE training up to 1T parameters, and FSDP throughput surpassing traditional 3D parallel schemes for MoE above 200B.