MiniMax-M1-40k

Training compute

4.2×10²⁴ FLOP

Parameters

456B

Published

Jun 13, 2025

MiniMax-M1-40k is an AI model developed by MiniMax (China), first published in June 2025. It works in the language domain, on tasks such as language modeling/generation, question answering, code generation and quantitative reasoning.

Training it took an estimated 4.2×10²⁴ FLOP of compute. The model has 456,000,000,000 parameters. It was trained on roughly 7.5T datapoints. Training ran on NVIDIA H800 SXM5.

Access: Open weights (unrestricted). Its weights are openly available. It is built on top of MiniMax-Text-01. Epoch AI rates the confidence of this record as likely.

Full record

Organization: MiniMax
Country of organization: China
Domain: Language
Task: Language modeling/generation, Question answering, Code generation, Quantitative reasoning
Training compute: 4.2×10²⁴ FLOP
Parameters: 456,000,000,000
Dataset size: 7.5T
Training hardware: NVIDIA H800 SXM5
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Base model: MiniMax-Text-01
Epoch confidence: Likely