525 articles
Llama 4
Llama 4 is a generation of open-weight large language models released by Meta on April 5, 2025, its first with mixture-of-experts architecture and native multimodality, shipping Scout and Maverick.
Llama Community License
The Llama Community License is Meta Platforms' source-available license for distributing Llama model weights, permitting commercial use while imposing restrictions like a 700 million monthly active user gate.
LLaVA
LLaVA (Large Language and Vision Assistant) is an open-source family of vision-language models, first released in April 2023, that connects a CLIP vision encoder to a language model.
LongCat (coding models)
LongCat is a family of open-weight mixture-of-experts large language models released by the Chinese technology company Meituan beginning in August 2025, positioned for repository-level coding and agentic tool use.
LongCat-Video
LongCat-Video is a 13.6-billion-parameter open-source video generation model released by Meituan in October 2025, known for producing minutes-long videos through iterative continuation and released under the MIT License.
Longformer
Longformer is a transformer language model introduced in 2020 by AI2 researchers for long documents, replacing standard self-attention with a linear-cost mix of local and global attention.
LongVie
LongVie is an open framework for controllable ultra-long video generation, released in August 2025 by NVIDIA and partners, with LongVie 2 following in December 2025 and claiming five-minute generation.
LTX Video
LTX Video is a family of open-weights, diffusion-transformer video generation models by Lightricks. It supports text, image, and audio-video generation at up to 4K/50 FPS.
Luma Dream Machine
Luma Dream Machine is a generative video product from San Francisco startup Luma AI, launched in June 2024 as a text-to-video tool and since rebuilt around the Ray model family.
Luminous (model family)
Luminous was a family of large language models developed by the German company Aleph Alpha, released from April 2022 and positioned as a step toward Europe's AI sovereignty.
Lyria
Lyria is a family of AI music generation models developed by Google DeepMind, first announced in November 2023 with YouTube and powering Dream Track and MusicFX DJ.
MAGI-2
MAGI-2 is an open-weight video generation model released by Beijing-based Sand AI on August 5, 2026, generating 10-second clips with synchronized audio from text or image prompts.
Magistral (Mistral AI)
Magistral is a family of reasoning language models developed by Mistral AI, announced in June 2025 with an open-weight Magistral Small and a proprietary Magistral Medium.
Magma (multimodal agent foundation model)
Magma is an 8-billion-parameter multimodal foundation model for AI agents released by Microsoft Research in February 2025, combining vision-language understanding with action-taking in digital interfaces and robotic manipulation.
MAI-Code-1-Flash
MAI-Code-1-Flash is a small coding model developed in-house by Microsoft AI and released in June 2026, Microsoft's first in-house coding model for GitHub Copilot, built to reduce Microsoft's dependence on OpenAI.
Make-A-Video
Make-A-Video is a text-to-video generation system built by Uriel Singer and colleagues at Meta AI, announced in September 2022, producing brief silent clips from text prompts.
Marble
Marble is a multimodal world model by World Labs (San Francisco). Founded by Fei-Fei Li, it is the first commercial AI to generate navigable 3D environments from diverse data.
Matrix-Game
Matrix-Game is a series of open interactive world foundation models from Skywork AI that generate playable game environments from a reference image and user actions, first released in 2025.
Matrix-Game 2.0
Matrix-Game 2.0 is an open-source interactive world model released by Skywork AI in 2025, streaming navigable video at 25 frames per second under an MIT license.
MatterSim
MatterSim is a machine learning force field from Microsoft Research AI for Science, introduced in 2024, covering the periodic table from 0 to 5000 K.
MBPP
MBPP (Mostly Basic Programming Problems) is a benchmark of 974 crowd-sourced entry-level Python tasks, released by Google Research in August 2021, that became a standard test for code-generating language models.
Med-Gemini
Med-Gemini is a family of multimodal medical AI models from Google Research and Google DeepMind, fine-tuned from Gemini on de-identified medical data and introduced in May 2024.
Med-PaLM
Med-PaLM is a family of large language models from Google Research fine-tuned for medical question answering; in 2023 it was the first model to pass USMLE-style questions.
MedSAM
MedSAM is a promptable AI foundation model for medical image segmentation, fine-tuned from Meta's Segment Anything Model on over 1.5 million medical image-mask pairs and introduced in April 2023.
Megatron-Turing NLG
Megatron-Turing NLG (MT-NLG) is a 530-billion-parameter language model trained jointly by Microsoft and NVIDIA and announced in October 2021, then the largest dense transformer model.
MelGAN
MelGAN is a non-autoregressive GAN vocoder introduced in October 2019 by Kundan Kumar and colleagues at Descript, converting mel-spectrograms into raw audio waveforms over 100 times faster than real time.
Mercury (diffusion LLM)
Mercury is a family of diffusion-based large language models from Inception Labs, launched in February 2025 with Mercury Coder Mini and Small, described as the first commercial-scale diffusion LLMs.
Meta's retreat from open weights
Meta's retreat from open weights was the 2025–26 shift in which Meta, long the standard-bearer for open Llama models, launched the closed, API-only Muse Spark, then partially reopened with Muse Glimmer in 2026.
Microsoft Azure Neural TTS
Microsoft Azure Neural TTS is the neural text-to-speech service in Azure AI Speech, converting text into natural speech with prebuilt, custom, and personal voices in 100+ languages.
Midjourney
Midjourney, also known as Midjourney.com, is a generative AI service creating images and video from text prompts, founded by David Holz in San Francisco in 2021.