Skip to content
#9 - MiniMax-01: Scaling Foundation Models with Lightning Attention episode artwork

Feb 5, 2025 31 min

#9 - MiniMax-01: Scaling Foundation Models with Lightning Attention

from Artificially Speaking Hosted by Henry Moran

The Paper details the development and capabilities of MiniMax-01, a series of large language models (LLMs) and vision-language models (VLMs). Key features include superior long-context processing (up to 4 million tokens), utilization of lightning attention, and a Mixture of Experts (MoE) architecture. Performance is shown to be comparable to state-of-the-art models like GPT-4 and Claude. The paper explores architectural design, computational optimizations, pre-training methodologies, and safety alignment. Finally, the models' code and weights are publicly released.

Episode details come from the podcast's RSS feed.

Embed this episode

Ready to play

#9 - MiniMax-01: Scaling Foundation Models with Lightning Attention

0:00 31:02

No transcript yet.

Requests help us decide what to transcribe next.

Speaking to Politics Mike Politics shape our world. The issues being decided in the halls of government every day have real and lasting impact on all our lives, but discussing politics is extremely unpleasant. It is very difficult today to have a calm, adult conversation about politics. Are the people screaming on TV, the radio, and online really who “we” are? Are they really representative of mainstream America or are they just really aggressive and loud? I believe the adults just got tired of it and walked away, but their silence is not consent. This podcast is the test of that theory. If you see fit to tune in, I will present news of the day, analysis and commentary while doing my best to check my feelings at the door and let you, the listeners, decide what it means to you. No ominous music, no angry rants, just a relaxed conversation and positive vibe. Politics, for reasonable people. Ukraine - History in the Making Emil Juhler When I started the conversation series ”Ukraine - History in the Making” I could have never imagined where this journey would lead me. It all started in March 2023 after I got home from an exchange stay in Chile and felt, that I had to do more for Ukraine, than what I did on my social media accounts and on the streets in the spring and summer of 2022. I think most of us remember that initial period after Ruzzia’s the full-scale invasion of Ukraine. Speaking from a European/Danish point of view it was a time of big fear, desperation and uncertainty, but also a time of high hopes, energy and big solidarity. In the very first conversation of the show with Veronika Netrebenko, she says the following about the feelings of Ukrainians during the Maidan Revolution: ”We understood we gotta fight for it, and it was something totally different, we felt different, we felt that we need to fight for it”. To a certain degree I believe you can say the same about what happened within the hearts and mi

Frequently asked questions

How long is this episode of Artificially Speaking?

This episode is 31 minutes long.

When was this Artificially Speaking episode published?

This episode was published on February 5, 2025.

Can I download this Artificially Speaking episode?

Yes. Use the download control on the episode player to save the publisher-provided media file.