# MAX Nightly 26.4.0.dev2026053016 (Mojo 1.0.0b2.dev2026053016) Released

**URL:** <https://forum.modular.com/t/max-nightly-26-4-0-dev2026053016-mojo-1-0-0b2-dev2026053016-released/3124>\
**Category:** Nightly\
**Created:** [May 30, 2026, 6:13pm UTC](https://forum.modular.com/t/max-nightly-26-4-0-dev2026053016-mojo-1-0-0b2-dev2026053016-released/3124 "2026-05-30T18:13:28Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Modular](https://sea1.discourse-cdn.com/flex001/user_avatar/forum.modular.com/modular/32/17_2.png) [@Modular](https://forum.modular.com/u/Modular)\
**Post date:** [May 30, 2026, 6:13pm UTC](https://forum.modular.com/t/max-nightly-26-4-0-dev2026053016-mojo-1-0-0b2-dev2026053016-released/3124/1 "2026-05-30T18:13:28Z")

</div>

🧑‍🚀 A new nightly version has been released! 🧑‍🚀

See the quickstart guide for installation instructions: [Quickstart | Modular](https://docs.modular.com/max/get-started/#set-up-your-project)

MAX changelog updates:

- Kimi K2.5 tool calling now supports interleaved thinking: a single  
assistant turn may interleave multiple `<think>...</think>` reasoning  
blocks with multiple tool-call sections and end with `<|im_end|>`. The  
constrained-decoding grammar (used for `tool_choice` and JSON  
`response_format`) admits up to eight tool-call sections with an optional  
reasoning block before each, and lets the model stop before the cap. This  
fixes a `tool_choice=auto` failure where a second tool-call section  
disabled grammar enforcement for the rest of the request.

Mojo changelog updates:

Raw MAX diff: [https://github.com/modular/modular/compare/d6c275f1d961a53d80a84d495180a2b1123ad2c6...c807ae59ddb6de5bd46241cbfbfb220e0a6fce46](https://github.com/modular/modular/compare/d6c275f1d961a53d80a84d495180a2b1123ad2c6...c807ae59ddb6de5bd46241cbfbfb220e0a6fce46))\>  
Current Mojo changelog: [https://github.com/modular/modular/blob/main/mojo/docs/nightly-changelog.md](https://github.com/modular/modular/blob/main/mojo/docs/nightly-changelog.md)