Meta released Muse Spark Open on October 15, 2026, describing it as the most capable U.S. open-weight AI model to date

Overview Debates Related claims Topics Sources

Chinese AI company MiniMax released MiniMax M3.1 Flash on October 19, 2026, featuring 428 billion total parameters, about 23 billion active parameters in a mixture-of-experts architecture, and a 1,000,000-token context window.

MiniMax reported that MiniMax M3.1 Flash operates about three times faster than MiniMax M3, with median output speeds consistently exceeding 170 tokens per second.

MiniMax M3.1 Flash API pricing is set at $0.15 per million input tokens and $0.60 per million output tokens, with cached input reads priced at $0.015 per million tokens.

MiniMax scheduled the open-weight model release for MiniMax M3.1 Flash for October 23, 2026.

MiniMax confirmed that an earlier build of MiniMax M3.1 Flash operated anonymously under the name Space Bunny, processing over 200 trillion tokens prior to its official launch.

Meta Platforms Inc. released its Muse Wonder foundation AI model and offers paid subscription tiers ranging from $20 to $200 per month for power users and developers.

China's proposed export controls announced on October 16, 2026 offer simplified licensing for open-source technology exports and Chinese companies operating their own computing centres overseas

Artificial intelligence startup DeepSeek successfully completed Chinese state registration, open-sourcing, and authorization procedures for overseas deployment of its AI models on October 22, 2026

MiniMax Group Inc. completed the registration process and authorization procedures with Chinese regulators under China's October 16, 2026 proposed export controls for open-sourcing and overseas deployment of its AI models on October 22, 2026

China Central Television affiliate Yuyuantantian reported that much of Chinese AI technology will remain free for export despite proposed export control rules

MiniMax reported that MiniMax M3.1 Flash can reach peak generation speeds exceeding 200 tokens per second.

Chinese state media account Yuyuan Tantian reported on October 23, 2026 that overseas internet users urged other Chinese open-source large language model developers to register for global use following DeepSeek's approval for overseas deployment of its AI models

DeepSeek plans to use the $15 billion raised in October 2026 to expand artificial intelligence data centers and develop open-source models for Chinese and international markets

DeepSeek released its open-source V4.1 Flash model in September 2026, causing share prices of rival domestic artificial intelligence firms MiniMax Group Inc. and Z.AI Co. to decline

The RFD4-Proteina model co-developed by Nvidia and the University of Washington generates both protein sequences and three-dimensional structures.

The UW Medicine Institute for Protein Design and Nvidia released RFD4-Proteina, an open-source AI protein design model that combines programmable protein design with search capabilities.

Unknown attackers compromised the LiteLLM open-source AI gateway release process on October 21, 2026, distributing malicious version 1.107.0 through PyPI and Docker registries.

LiteLLM version 1.106.2 is the latest safe release following the October 21, 2026 supply chain compromise.

PyPI quarantined LiteLLM version 1.107.0 within hours of publication, and Docker images were removed from Docker Hub and GitHub Container Registry.

LiteLLM, developed by BerriAI, provides a shared interface for LLM providers including OpenAI, Anthropic, and Google Gemini, receiving approximately 23 million weekly downloads.