MiniMaxText01

Model type	MiniMaxText01
Class	MiniMaxCausalLMModel
Task	hybrid-text-generation
Source	models/minimax.py

Description

MiniMax causal language model with hybrid Lightning + GQA + MoE.

Architecture:

• config.layer_types preserves the checkpoint’s explicit per-layer schedule

• Full-attention layers use partial-RoPE GQA

• Lightning layers use a fixed-size recurrent state and no KV cache

Lightning Attention layers carry a single recurrent_state tensor of shape (B, num_heads, head_dim, head_dim) per layer. Full attention layers use the standard KV cache.

Task: hybrid-text-generation (HybridCausalLMTask).

HuggingFace model types: "MiniMaxText01" and "minimax"

Usage

mobius build --model <MODEL_ID> output_dir/

from mobius import build

model = build("<MODEL_ID>")