Skip to content
trckr.ai

Loading…

The archive

All releases

Every model release from the labs trckr follows, newest first: spotted live since November 2026, and from the archive back to 2018.

RSS for this view

November 2026

MetaText & chat

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

Muse Ember

Meta's new small, low-cost model for running AI agents that carry out tasks, priced to undercut rivals for developers.

MeituanText & chat

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.OpenOpen model: Its files are public (open weights), so anyone can download it and run it on their own computer, usually for free.

LongCat-2.5

Meituan, the Chinese services company, released LongCat-2.5, a free-to-download AI assistant that can read images and handle long coding and computer tasks.

NVIDIASpecialist

MinorMinor update: A small change to an existing model: a dated version, a preview refresh or a variant.OpenOpen model: Its files are public (open weights), so anyone can download it and run it on their own computer, usually for free.

Video-to-Data GR00T N1.7 Tissue-Box Lift

NVIDIA's robot-training example, free to download, that teaches a simulated robot to lift, hold and put back a tissue box. Made for robotics researchers.

TencentText & chat

MinorMinor update: A small change to an existing model: a dated version, a preview refresh or a variant.OpenOpen model: Its files are public (open weights), so anyone can download it and run it on their own computer, usually for free.

Qwen3.8-27B-DSpine

Tencent released a free-to-download helper model that speeds up Alibaba's Qwen3.8-27B by drafting text in parallel, mainly for developers running AI at scale.

GoogleVideo

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

Gemini Omni 1.2 Flash

Google's new video maker and editor, which can turn text or pictures into short clips up to 15 seconds long.

OpenAIText & chat

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

GPT-6-Cyber

OpenAI's new AI for security teams, built to help find software weaknesses and check fixes in approved test settings.

BaiduText & chat

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.OpenOpen model: Its files are public (open weights), so anyone can download it and run it on their own computer, usually for free.

ERNIE 5.5 Flash-Base

Baidu's large new model for developers and researchers, free to download; it is a starting point for building their own AI tools and needs powerful servers to run.

OpenAIImage

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

GPT-Image-2.6

OpenAI's new image maker that creates and edits pictures, offered in two versions to developers.

Z.aiText & chat

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

GLM-5.3-FlashX-Legal

A version of Zhipu's fast GLM model trained further for legal work, able to read text, pictures and video, offered for businesses and developers.

October 2026

GoogleText & chat

Not public yetNot public yet: It exists, but only testers can use it so far. trckr updates its page when it opens up.MajorMajor release: A new general-purpose AI model from a leading lab, the kind behind ChatGPT, Claude or Gemini: a new flagship or generation.

Gemini 4 Carbon

Google's most powerful AI model so far, built for cybersecurity work and long, complicated tasks. For now only security teams in a Google programme can use it; others get it later, with no date yet.

Liquid AIText & chat

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.OpenOpen model: Its files are public (open weights), so anyone can download it and run it on their own computer, usually for free.

LFM3-1.7B

A small model from Liquid AI, free to download and light enough for a phone or laptop, made for building AI agents that use tools and produce structured answers.

GoogleVideo

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

Genie 4

Google's Genie 4 makes videos and sound, turns still images into video, and builds 3D scenes, with access limited to select users.

MicrosoftImage

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

MAI-Image-2.7

Microsoft's new image model creates pictures from text and edits existing images. It is available online only and is not free to download.

MicrosoftText & chat

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

MAI-Cyber-1.1-Flash

Microsoft's MAI Cyber 1.1 Flash is a model for finding and exploiting software security flaws and for coding, available online only.

OpenAIText & chat

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

GPT-6.1 Terra

OpenAI's GPT-6.1 Terra is a text model for writing and answering questions, available to developers building apps and not free to download.

Moonshot AIText & chat

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.OpenOpen model: Its files are public (open weights), so anyone can download it and run it on their own computer, usually for free.

Kimi K3.1 Pro

Moonshot AI's Kimi K3.1 Pro is a text model for writing and answering questions, and its files are free to download, with licence limits only for very large companies.

Mistral AIVoice & audio

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

Voxtral 3

Mistral AI's new family of tools for turning speech into text and text into spoken audio, offered to developers.

MiniMaxText & chat

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

MiniMax M3.1 Flash

Chinese AI company MiniMax has released a quick, lighter assistant that can read pictures and handle coding and multi-step tasks, available through OpenRouter.

NVIDIAText & chat

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.OpenOpen model: Its files are public (open weights), so anyone can download it and run it on their own computer, usually for free.

Nemotron 3.5 Super VL

NVIDIA, the chip maker, released a free-to-download AI that can read text, pictures, documents and video, for developers building smart assistants.

OpenAIVoice & audio

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

GPT-Live-1.1 Quip

OpenAI's GPT-Live-1.1 Quip is a speech model, and details on how people can access it have not been published yet.

OpenAIVoice & audio

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

GPT-Live-1.1

OpenAI's new speech model for spoken conversations, not free to download.

KlingVideo

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

Kling 4.0

Kuaishou's Kling 4.0 makes videos from text or images, adds sound, and edits clips, available online only.

MetaText & chat

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.OpenOpen model: Its files are public (open weights), so anyone can download it and run it on their own computer, usually for free.

Muse Spark Open

Meta's Muse Spark Open is a text model for writing and answering questions that is free to download under restricted use terms.

xAIVoice & audio

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

Grok Voice 3.0

xAI's new voice model that turns spoken input into spoken replies, offered to developers and not free to download.

MetaText & chat

MajorMajor release: A new general-purpose AI model from a leading lab, the kind behind ChatGPT, Claude or Gemini: a new flagship or generation.

Muse Wonder Full

Meta's Muse Wonder Full model answers questions and understands images and video, offered to developers and not free to download.

MetaText & chat

MajorMajor release: A new general-purpose AI model from a leading lab, the kind behind ChatGPT, Claude or Gemini: a new flagship or generation.

Muse Wonder

Meta's Muse Wonder model answers questions and understands images and video, offered to developers and not free to download.

AnthropicText & chat

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

Claude Haiku 5.5

Anthropic's Claude Haiku 5.5 writes text and understands images, available to developers and not free to download.

GoogleImage

NotableNotable release: A meaningful new model: a new size or version, or a strong free-to-download model from a known lab.

Gemini 3.6 Flash Image (Nano Banana 2.1)

Google's new image model creates and edits pictures from text descriptions, available to developers and not free to download.

Free alerts

Know the moment a new model drops.

Pick one or several. Each has its own settings for what you hear about.