MiniMaxAI
PulseAugur coverage of MiniMaxAI — every cluster mentioning MiniMaxAI across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
FusionCow/Sulphur-3 model gains traction on Hugging Face
The FusionCow/Sulphur-3 model, based on MiniMaxAI/MiniMax-H3, has seen 195 downloads in the past month on Hugging Face. It is not currently deployed by any inference providers. The model's repository includes a README f…
-
MiniMax releases new MiniMax-M3.1-Flash model
MiniMax has released the MiniMax-M3.1-Flash model, which is available on Hugging Face. This release provides a new option for users working with local large language models.
-
StableDiffusion workflow enables video background replacement
A new workflow for StableDiffusion, dubbed Minimax H3, has been released, enabling the replacement of backgrounds in videos. This workflow splits long videos into overlapping regions, processes them simultaneously, and …
-
LLMs achieve massive context windows on consumer hardware with new techniques
Researchers are developing innovative methods to enable large language models (LLMs) to handle significantly larger context windows, even on consumer hardware. One approach, JustFit, uses techniques like KV compression …
-
MiniMax AI releases open-weights music generation model MiniMax-Music3
MiniMax AI has released MiniMax-Music3, a new open-weights music generation model. The model is described as production-ready and versatile, with a demo available for users to explore its capabilities. Links to the mode…
-
New AI Models Released: DeepSeek, LiquidAI, OMLAB, and MiniMaxAI on Hugging Face
Several AI models have been released and are available on Hugging Face. DeepSeek-V4-Pro-0813 is a model from DeepSeek. LiquidAI has released LFM2.5-VL-3B, a multimodal model designed for on-device deployment that can pr…
-
Open-weight AI models gain traction on Hugging Face
Several open-weight AI models are gaining traction on Hugging Face, with MiniMaxAI/MiniMax-H3 leading in downloads and stars for its image-text-to-video capabilities. Other models like DeepSeek-V4-Flash-0731, DavidAU/Qw…
-
JetBrains and MiniMaxAI release open-source models
JetBrains has released Mellum2-12B-A2.5B-Thinking, an open-source model with 12 billion parameters utilizing a Mixture of Experts architecture. Concurrently, MiniMaxAI's M3 model is now accessible through the Cloudflare…
-
MiniMaxAI releases multimodal M3 model with 1M context window
MiniMaxAI has released MiniMax-M3, a multimodal large language model with a 1 million token context window. This model boasts approximately 428 billion total parameters, with around 23 billion activated during operation…