rltakashige b315035ae0 Add minimax and fix qwen sharding strategies (#1318) ...

## Motivation

MiniMax tensor sharding does not provide equivalent outputs to running
it as a single node because RMSNorm weights cannot be split without
affecting the output.

Qwen3Next sharding was broken, and something with Qwen3MoE was likely
changed upstream, as several variables no longer exist.

This also ballooned into fixing prefix caching for non-standard models
as Qwen3Next was behaving weirdly.

## Changes

<!-- Describe what you changed in detail -->

## Why It Works

<!-- Explain why your approach solves the problem -->

## Test Plan

### Manual Testing
Worked for a 8 hour long eval at the same performance and a more similar
completion/reasoning token distribution.

---------

Co-authored-by: Alex Cheema <41707476+AlexCheema@users.noreply.github.com>
Co-authored-by: Alex Cheema <alexcheema123@gmail.com>
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
Co-authored-by: Evan <evanev7@gmail.com>

2026-02-06 13:26:59 +00:00

..

darwin-build-fixes.patch

mlx: build with Nix (#1285)

2026-01-29 14:07:00 +00:00

metal-toolchain.nix

mlx: build with Nix (#1285)

2026-01-29 14:07:00 +00:00

mlx.nix

Add minimax and fix qwen sharding strategies (#1318)

2026-02-06 13:26:59 +00:00