Articles Model tracker AI tools Learn AI About YouTube ↗ Instagram ↗

Updated daily

AI Model Release Tracker

Every new AI model we have verified, in one list. What it is, who made it and what is new, in plain English, with a link to the source.

83 models from 33 organisations since 3 June 2026. Last updated 21 November 2026.

Models tracked83
Organisations33
This month15
Last updated21 Nov
Releases over time, by organisation
Text Image Video Audio Code Science, robotics and other
JunJulAugSeptOctNov Google DeepMind Liquid AI OpenAI IBM Meta NVIDIA Anthropic Other organisations MAI-Code-1-Flash · Microsoft · 3 Jun Gemma 4 12B · Google DeepMind · 5 Jun Claude Fable 5 · Anthropic · 10 Jun Gemini 3.5 Live Translate · Google DeepMind · 10 Jun DiffusionGemma · Google DeepMind · 11 Jun Lyria 3 · Google DeepMind · 17 Jun GLM-5.2 · Zhipu AI · 21 Jun Seedance 2.5 · ByteDance · 25 Jun Base1 · Base44 · 30 Jun Claude Sonnet 5 · Anthropic · 1 Jul Nano Banana 2 Lite · Google DeepMind · 1 Jul GPT-Live-1 and GPT-Live-1 mini · OpenAI · 9 Jul Grok 4.5 · SpaceXAI · 9 Jul GPT-5.6 · OpenAI · 10 Jul GPT-5.6 Sol · OpenAI · 10 Jul Muse Spark · Meta · 10 Jul Muse Spark 1.1 · Meta · 12 Jul Inkling · Thinking Machines Lab · 16 Jul Orbital · Applied Computing · 16 Jul Inkling 2 Small · Thinking Machines · 19 Jul Cosmos 3 Edge · NVIDIA · 21 Jul Gemini 3.5 Flash Cyber · Google DeepMind · 22 Jul Gemini 3.6 Flash · Google DeepMind · 22 Jul GPT 5.6-Luna · OpenAI · 24 Jul ChatGPT-Live · OpenAI · 25 Jul Cosmos-H-Dreams · NVIDIA · 27 Jul Lyria 3.5 · Google DeepMind · 30 Jul Gemini Robotics 2 · Google DeepMind · 31 Jul LFM2.5-Encoder-230M and LFM2.5-Encoder-350M · Liquid AI · 2 Aug OlmoEarth · AI2 · 2 Aug LFM2.5-2.6B · Liquid AI · 5 Aug Kimi K3 · Moonshot AI · 8 Aug Muse Glimmer · Meta · 11 Aug Gemini 3.7 Flash · Google DeepMind · 14 Aug Gemini Robotics ER 2 · Google DeepMind · 15 Aug LFM2.5-VL-3B · Liquid AI · 15 Aug Palmyra X6 · Writer · 16 Aug Magpie Multilingual TTS · NVIDIA · 21 Aug GLM-5.3 · Z.ai · 23 Aug Gemini 3.5 Transcribe · Google DeepMind · 27 Aug Gemini Omni 1.1 Flash · Google DeepMind · 27 Aug Granite 4.2 · IBM · 27 Aug Gemini 3.8 Flash · Google DeepMind · 3 Sept GPT-6 Astra · OpenAI · 4 Sept WeatherNext 3 · Google DeepMind · 4 Sept Granite Time Series PatchTST-FM-r2 · IBM · 8 Sept Suno v6 · Suno · 9 Sept Koa · Salesforce · 15 Sept Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking · Google DeepMind · 16 Sept Jev · TypeSafe AI · 19 Sept LFM2.5-VL-DSpark · Liquid AI · 25 Sept Gemini 3.8 Flash Cyber · Google DeepMind · 26 Sept Falcon-Emirati-7B · Technology Innovation Institute · 6 Oct EmbeddingGemma 2 · Google DeepMind · 7 Oct Falcon-ASR · Technology Innovation Institute · 10 Oct LFM2.5-24B-A2B · Liquid AI · 15 Oct Beam · Reflection · 17 Oct Gemini 4 Argon · Google DeepMind · 17 Oct Granite Switch 4.1 · IBM · 18 Oct FLUX.3 Image Klein · Black Forest Labs · 20 Oct Voxtral 3 · Mistral · 20 Oct Carbon-X · Hugging Face · 22 Oct Flint · Recursive · 22 Oct Canva Craft · Canva · 23 Oct Granite Speech 5.0 2B · IBM · 24 Oct Dessa · Block · 25 Oct Genie 4 · Google DeepMind · 28 Oct LFM3 · Liquid AI · 30 Oct Gemini 4 Carbon · Google DeepMind · 1 Nov Mistral Large 4 · Mistral · 1 Nov Muse Realtime Avatar and Muse Realtime Voice · Meta · 6 Nov Gemini Omni 1.2 Flash · Google DeepMind · 7 Nov Phonon · Hugging Face · 7 Nov Cosmos 3 Nano · NVIDIA · 8 Nov Dharma Field Extractor · Dharma AI · 8 Nov Claude Fable 5.2 and Claude Mythos 5.2 · Anthropic · 9 Nov AlphaProof Code · Google DeepMind · 10 Nov BotPsy · Tether AI · 11 Nov AlphaGenome 2 · Google DeepMind · 12 Nov Timestral · Mistral · 15 Nov Gemini 4 Hydrogen · Google DeepMind · 16 Nov Rumble-2-Flash · Reson8 · 19 Nov Gemini 4 Neon · Google DeepMind · 20 Nov

Each mark is one release. Hover or tap a mark for its name, or select it to jump to the entry below. Organisations with a single release share the bottom row.

November 2026

Gemini 4 Neon Google DeepMind Text

Frontier-level intelligence model for software engineering, knowledge work, and science workflows with fast task completion.

Gemini API, Google AI Studio, Gemini for business, Google AI Pro and Ultra subscribers Source: Google DeepMind ↗

Rumble-2-Flash Reson8 Audio

0.5B-parameter speech recognition model for 40+ languages with automatic language detection.

Open source under Apache 2.0 license, available on Hugging Face Hub and via API. Source: HuggingFace Blog ↗

Gemini 4 Hydrogen Google DeepMind Text

Fast execution model for everyday coding and agentic workflows, available across Google AI products and services.

Google AI Studio, Gemini API, Gemini for business, Gemini app, Google Search, Google AI Pro and Ultra subscribers Source: Google DeepMind ↗

Timestral Mistral Other

Family of time-series foundation models in four sizes from 5M to 209M parameters for probabilistic multivariate forecasting.

Available as a family of four models; fine-tuning available via Mistral Forge. Source: Mistral AI ↗

AlphaGenome 2 Google DeepMind Science

Next-generation DNA sequence model predicting how genetic variants affect specific human cell types.

Non-commercial research via API; private preview for commercial research Source: Google DeepMind ↗

BotPsy Tether AI Text

A 0.8B-parameter model for on-device function calling and agentic workflows built on Qwen3.5-0.8B.

Model safetensors and GGUFs available for download Source: HuggingFace Blog ↗

AlphaProof Code Google DeepMind Code

AI framework for formal verification that mathematically proves software cryptographic guarantees and security properties.

Researchers released the specification and proof for the Signal protocol analysis Source: Wired AI ↗

Claude Fable 5.2 and Claude Mythos 5.2 Anthropic Code

Upgraded models with improvements in coding, reasoning, and vision capabilities, generating outputs up to 30% faster than previous versions.

Fable 5.2 available to all users. Mythos 5.2 requires application through Life Sciences and Cyber Verification Programs. Source: Anthropic ↗

Dharma Field Extractor Dharma AI Other

A 2-billion-parameter model that extracts structured JSON fields from any document type.

Source: HuggingFace Blog ↗

Cosmos 3 Nano NVIDIA Video

Four open-weight checkpoints for interactive world simulation in driving and robotics tasks.

Open weights under OpenMDW1.1 for commercial and non-commercial use. Source: HuggingFace Blog ↗

Phonon Hugging Face Audio

On-device text-to-speech model running locally without server or network connection.

Phonon 31m Apache 2.0 free for all uses. Phonon 40m CC BY-NC free for non-commercial. Source: HuggingFace Blog ↗

Gemini Omni 1.2 Flash Google DeepMind Video

Generates native 15-second videos with improved consistency, multi-turn editing, and 30% lower costs than Omni 1.1.

Generally available on Gemini API, Google Cloud, Google Vids, and Gemini for business Source: Google DeepMind ↗

Muse Realtime Avatar and Muse Realtime Voice Meta Other

New AI models powering Muse's digital avatar for video chat and voice call capabilities with animated character expressions.

Available to all U.S. users on mobile, web, and Mac Source: TechCrunch AI ↗

Mistral Large 4 Mistral Text

Multimodal, multilingual Mixture-of-Experts model with 1T total parameters and 1M token context window.

Released under Mistral Ecosystem Open-Weight License on Hugging Face. Source: HuggingFace Blog ↗

Gemini 4 Carbon Google DeepMind Text

Model for coding and agentic tasks, with a focus on cybersecurity and long-running work.

Starting this week, subset of cyber defenders; gradual access through Fairwind Program. Source: Google DeepMind ↗

October 2026

LFM3 Liquid AI Other

New generation of Liquid Foundation Models with text, vision, audio, omni, and decision models in 2B parameter range.

Available via Hugging Face transformers library for on-device deployment Source: HuggingFace Blog ↗

Genie 4 Google DeepMind Video

World model generating photorealistic interactive worlds with real-time audio, free-form editing, and extended scene consistency.

Research preview for trusted testers; rolling out to Google AI Ultra subscribers and early access partners Source: TechCrunch AI ↗

Dessa Block Audio

Full-duplex speech-to-speech model trained entirely on speech that can run locally on consumer GPUs.

No public access currently; Block seeking safety researchers to test the model. Source: TechCrunch AI ↗

Granite Speech 5.0 2B IBM Audio

New multilingual speech model for transcription and translation with improved accuracy and robustness.

Apache 2.0 license with support in Transformers, vLLM, llama.cpp, and mlx-audio Source: HuggingFace Blog ↗

Canva Craft Canva Image

Image generation model that interprets prompts and generates images up to 2K resolution with accurate text and details.

Available to all Canva platform users via Dream Lab tab and Canva API Source: TechCrunch AI ↗

Flint Recursive Code

A coding model that runs on laptops and was built largely with AI assistance from the Eureka Machine.

Free download on Hugging Face under Apache 2.0 open source license Source: TechCrunch AI ↗

Carbon-X Hugging Face Science

An open model that enables conversational queries about DNA sequences and their effects on gene activity and health.

Open model that runs locally on a laptop without sending data to Hugging Face Source: Wired AI ↗

Voxtral 3 Mistral Audio

Speech-to-text and text-to-speech models with voice agent SDK supporting 50+ languages for real-time conversations.

Available via Mistral's EU-hosted API or for self-deployment on own infrastructure Source: Mistral AI ↗

FLUX.3 Image Klein Black Forest Labs Image

Open image generation model supporting text-to-image, image-to-image, and bounding box layout control.

Open checkpoints available, runs in Diffusers, accessible via Hugging Face space Source: HuggingFace Blog ↗

Granite Switch 4.1 IBM Text

Modular models combining chat, decision-making, and speech-to-text capabilities in one endpoint.

Available in 8B and 30B sizes, runs in vLLM, llama.cpp, and downstream platforms. Source: HuggingFace Blog ↗

Gemini 4 Argon Google DeepMind Robotics

A physical reasoning model for robotics that scores 87% on the Embodied Reasoning benchmark, outperforming other frontier models.

Source: Google DeepMind ↗

Beam Reflection Text

501B parameter mixture-of-experts model with 23B active parameters per token, supporting 1M token context window.

Open source under Apache 2.0 license on Hugging Face Source: HuggingFace Blog ↗

LFM2.5-24B-A2B Liquid AI Text

A 24B parameter reasoning model with 2B active parameters designed for fast on-device agents.

Open-weight model available on Hugging Face Source: HuggingFace Blog ↗

Falcon-ASR Technology Innovation Institute Audio

1.6 billion parameter speech recognition model for Arabic with focus on Emirati dialect, also supporting English, French, Spanish and Portuguese.

Hugging Face Demo Space available; API access and native applications planned Source: HuggingFace Blog ↗

EmbeddingGemma 2 Google DeepMind Other

Multimodal embedding model mapping text, images, audio, and video into unified embedding space with 740 million parameters.

Apache 2.0 license, available on Hugging Face and Kaggle Source: Google DeepMind ↗

Falcon-Emirati-7B Technology Innovation Institute Text

A dialect-specialized model built to understand and generate Emirati Arabic with cultural context and nuance.

Source: HuggingFace Blog ↗

September 2026

Gemini 3.8 Flash Cyber Google DeepMind Code

Specialized cyber defense model that autonomously finds and fixes vulnerabilities in code at reduced operating cost.

Limited access through Fairwind Program for governments, critical infrastructure operators, and trusted enterprise partners Source: Google DeepMind ↗

LFM2.5-VL-DSpark Liquid AI Other

Experimental DSpark draft model for vision-language model LFM2.5-VL-3B enabling faster inference through speculative decoding.

Open-weight. Available on Hugging Face in Safetensors and GGUF formats. Source: HuggingFace Blog ↗

Jev TypeSafe AI Other

A transformer-based model that produces calibrated decisions instead of text, designed for software automation tasks.

Available via API; developers can access it through TypeSafe AI's service. Source: TechCrunch AI ↗

Gemini 3.8 Live and Gemini 3.8 Live Extended Thinking Google DeepMind Audio

Two advanced voice dialogue models with improved reasoning, real-time visual processing, and background task execution capabilities.

Available today via Gemini API, Google Workspace, and Gemini app Source: Google DeepMind ↗

Koa Salesforce Text

Reasoning model built on Nvidia's Nemotron, trained for sales, marketing, and customer support tasks.

Provided as alternative in Salesforce's Agentforce platform for enterprise customers. Source: TechCrunch AI ↗

Suno v6 Suno Audio

New model family trained on licensed music data from major music labels and distributors.

Base and wild versions for paying users, mini version for all users Source: TechCrunch AI ↗

Granite Time Series PatchTST-FM-r2 IBM Other

Time series forecasting model with conformer architecture, larger pretraining corpus, probabilistic forecasting, and missing value imputation.

Open weights and code available under Apache 2.0 and OpenMDW 1.0 licenses Source: HuggingFace Blog ↗

WeatherNext 3 Google DeepMind Science

AI weather model using real-time satellite data for hourly high-resolution forecasts five times sharper than previous version.

Integrated into Google Search, Gemini, Maps, and Google Cloud Platform. Source: Google DeepMind ↗

GPT-6 Astra OpenAI Other

Navigates computers and web browsers, writes software, and solves difficult math problems.

Paid customers starting with Daybreak early access program, then ChatGPT Plus, Pro, Business and Enterprise subscribers Source: Wired AI ↗

Gemini 3.8 Flash Google DeepMind Text

Google released Gemini 3.8 Flash with improved reasoning and coding capabilities, available in standard and Cyber variants.

Standard version available across Google ecosystem; Cyber version limited to trusted testers and governments Source: Ars Technica AI ↗

August 2026

Granite 4.2 IBM Text

Open-weight models in 3B, 8B, and 30B variants with 128,000-token context window and reasoning capabilities.

Open-weight models designed to be downloaded and self-hosted. Source: Ars Technica AI ↗

Gemini Omni 1.1 Flash Google DeepMind Video

Production-ready update enabling developers to extend scenes, control frame transitions, and generate 4K video with improved visual consistency.

Available via Google AI Studio and Gemini API for developers Source: Google DeepMind ↗

Gemini 3.5 Transcribe Google DeepMind Audio

Speech-to-text model for real-time transcription with improved word error rates, multi-speaker identification, and support for over 85 languages.

Available via Gemini API, Google AI Studio, Gemini Enterprise Agent Platform, and integrated into Gboard, Gemini app, and Chrome. Source: Google DeepMind ↗

GLM-5.3 Z.ai Text

Extended post-training version of GLM-5.2 base model with improved benchmark performance on coding tasks.

Currently in coding plan, coming to API and Hugging Face open weights in two weeks Source: Interconnects ↗

Magpie Multilingual TTS NVIDIA Audio

Multilingual TTS model expanded to 12 languages including Modern Standard Arabic, Korean, and Brazilian Portuguese with improved quality.

Open weights on Hugging Face and production-ready NVIDIA NIM for own infrastructure deployment. Source: HuggingFace Blog ↗

Palmyra X6 Writer Text

Post-training variation on GLM-5.2 designed to reduce deployment costs by up to 50% for basic tasks.

Available to Writer clients starting Thursday Source: TechCrunch AI ↗

LFM2.5-VL-3B Liquid AI Image

Vision-language model with improved screen understanding, grounding, multi-image reasoning, and function calling for edge devices.

Available on Hugging Face. Can run on-device or via GPU inference. Source: HuggingFace Blog ↗

Gemini Robotics ER 2 Google DeepMind Robotics

A robotics model enabling real-time spatial reasoning, multi-step task planning, and multi-robot collaboration with video understanding.

Publicly available via Gemini API, Google AI Studio, and Gemini Enterprise Agent Platform private preview Source: Google DeepMind ↗

Gemini 3.7 Flash Google DeepMind Code

Improved coding and agent model with better performance on debugging, code generation, and web development tasks.

Gemini API, Google AI Studio, Android Studio, Gemini Enterprise, Spark for Pro and Ultra subscribers Source: Google DeepMind ↗

Muse Glimmer Meta Other

Multimodal 30B parameter model distilled from Muse, designed for local agentic use cases with vision and language capabilities.

Open source under Apache 2.0 license, available on Hugging Face Hub for local deployment. Source: HuggingFace Blog ↗

Kimi K3 Moonshot AI Text

Chinese language model released with large context window and improved capabilities across multiple domains.

Available via API with subscription plans of $40, $100, and $200 per month Source: Interconnects ↗

LFM2.5-2.6B Liquid AI Text

A 2.6 billion parameter model designed for on-device agent deployment with tool calling and multi-step workflows.

Available on Hugging Face for download and use Source: HuggingFace Blog ↗

OlmoEarth AI2 Science

Family of Earth observation foundation models pretrained on roughly 10 terabytes of multimodal satellite data.

Governments, NGOs, and mission-driven organizations can adapt OlmoEarth for applications. Source: HuggingFace Blog ↗

LFM2.5-Encoder-230M and LFM2.5-Encoder-350M Liquid AI Text

Two encoder models matching larger models' quality while supporting 8,192-token context with fast CPU inference.

Available on Hugging Face for download and fine-tuning. Source: HuggingFace Blog ↗

July 2026

Gemini Robotics 2 Google DeepMind Robotics

AI model combining vision language and action models to control robots performing complex physical tasks.

Source: Wired AI ↗

Lyria 3.5 Google DeepMind Audio

Music generation model with improved musicality, lyrics, vocals, and creative control features.

Rolling out today in Google Flow Music Source: Google DeepMind ↗

Cosmos-H-Dreams NVIDIA Robotics

Real-time action-conditioned generative simulator for surgical robotics that distills Cosmos-H-Surgical-Simulator for interactive closed-loop control.

Runs on single NVIDIA RTX PRO 6000 GPU; integrated with dVRK and Versius platforms. Source: HuggingFace Blog ↗

ChatGPT-Live OpenAI Audio

New family of voice models enabling ChatGPT Voice to control AI agents and perform tasks on computers.

Rolling out globally in ChatGPT desktop app and iOS app via remote access Source: TechCrunch AI ↗

GPT 5.6-Luna OpenAI Text

Smallest model from latest release outperforms GPT 5.5 on HealthBench evaluation for health queries.

Logged-in U.S. users with free, Go, Plus, Pro plans on web and iOS Source: TechCrunch AI ↗

Gemini 3.6 Flash Google DeepMind Text

More efficient than 3.5 Flash with better coding, knowledge work, and multimodal performance at lower cost.

Available via Gemini API and Gemini Enterprise Source: Google DeepMind ↗

Gemini 3.5 Flash Cyber Google DeepMind Code

Lightweight cybersecurity model fine-tuned to find, validate, and patch vulnerabilities quickly and efficiently.

Limited-access pilot via CodeMender, exclusively for governments and trusted partners. Source: Google DeepMind ↗

Cosmos 3 Edge NVIDIA Robotics

A 4-billion-parameter open world model for robots and vision AI agents to understand surroundings and generate actions on edge devices.

Open model available on Hugging Face for download and deployment on NVIDIA edge devices. Source: HuggingFace Blog ↗

Inkling 2 Small Thinking Machines Image

Second generation multimodal model with 594B total and 17B active parameters accepting text and image inputs.

Open model available on Hugging Face Hub Source: HuggingFace Blog ↗

Orbital Applied Computing Other

Foundation AI model combining time series, physics-based, and language models to predict oil and gas facility states.

In use at large publicly listed upstream oil and gas, downstream refining and petrochemicals companies Source: TechCrunch AI ↗

Inkling Thinking Machines Lab Text

Open-weight mixture-of-experts model with 975B parameters trained on multimodal data, outputs text and code.

Open-weight, downloadable and modifiable by outside developers and companies. Source: TechCrunch AI ↗

Muse Spark 1.1 Meta Text

Multimodal reasoning model with improvements in tool use, coding, and computer understanding for agentic tasks.

Public preview via Meta Model API and Meta AI app at meta.ai Source: AI at Meta ↗

Muse Spark Meta Text

New AI model better at coding, performing at or near levels of leading competitors, with paid version option.

Paid version available to customers at lower cost than competing models Source: The New York Times ↗

GPT-5.6 Sol OpenAI Text

OpenAI released GPT-5.6 Sol, its flagship artificial intelligence model with capabilities for office work and cybersecurity tasks.

Publicly released to individuals and businesses Source: The New York Times ↗

GPT-5.6 OpenAI Text

New model family with three variants offering improved efficiency, cost-effectiveness, and cybersecurity capabilities.

Available across ChatGPT, Codex, and the OpenAI API Source: TechCrunch AI ↗

Grok 4.5 SpaceXAI Text

SpaceXAI released Grok 4.5, a model with twice greater token efficiency than competitors, designed for coding, writing, research, and knowledge work.

Available to the public, with beta testing completed Source: TechCrunch AI ↗

GPT-Live-1 and GPT-Live-1 mini OpenAI Audio

Full-duplex voice models that sound more natural, handle turn-taking better, and allow users to interrupt naturally in conversations.

GPT-Live-1 mini replaces Advanced Voice Mode by default; GPT-Live-1 for paid tier users. Source: TechCrunch AI ↗

Nano Banana 2 Lite Google DeepMind Image

Fast and cheap image generation model that creates images in about 4 seconds at $0.034 per 1K images.

Google AI Studio, API, and Gemini by selecting Flash-Lite option Source: Ars Technica AI ↗

Claude Sonnet 5 Anthropic Text

More powerful agentic midsize model with improved reasoning, tool use, coding, and knowledge work capabilities.

Default for free and Pro plans, available for every subscription starting Tuesday Source: TechCrunch AI ↗

June 2026

Base1 Base44 Code

A custom LLM trained on tens of millions of real user interactions to support app creation with natural language.

Rolling out to Base44 users Source: TechCrunch AI ↗

Seedance 2.5 ByteDance Video

Generates 30-second 4K videos from a single prompt with up to 50 reference pieces for greater control.

Expected to launch in China next month; may be available July 9. Source: CNET ↗

GLM-5.2 Zhipu AI Code

Long-context model with 1M token context, improved coding capabilities, and IndexShare architecture optimization.

MIT open-source license, no regional limits Source: HuggingFace Blog ↗

Lyria 3 Google DeepMind Audio

Music generation model that creates music tracks from text prompts and images.

Available in Gemini app on Android 17 Pixel devices Source: TechCrunch AI ↗

DiffusionGemma Google DeepMind Text

A text generation model that produces outputs in parallel rather than sequentially, achieving 4x faster speeds on local hardware.

Available under Apache 2.0 license, downloadable from Hugging Face. Source: Ars Technica AI ↗

Gemini 3.5 Live Translate Google DeepMind Audio

Audio model delivering near real-time speech-to-speech translation in over 70 languages with natural intonation and continuous generation.

Developers via Gemini Live API, enterprises in Google Meet, everyone via Google Translate app. Source: Google DeepMind ↗

Claude Fable 5 Anthropic Text

First publicly available version of Mythos model with safety limits on high-risk areas.

Claude API, Enterprise plans, with staged rollout to subscriptions through June 22. Source: TechCrunch AI ↗

Gemma 4 12B Google DeepMind Text

A 12-billion-parameter model optimized to run on consumer laptops with 16GB of RAM.

Available for download on Kaggle and Hugging Face. Accessible via LM Studio and Google AI Edge Gallery. Source: Ars Technica AI ↗

MAI-Code-1-Flash Microsoft Code

AI coding model that converts written descriptions into source code for applications and websites.

Available in GitHub Copilot and Visual Studio Code text editor Source: CNBC ↗

How this list is made

Each morning our pipeline checks the day's fact-checked AI news for new model releases. An entry is added only when the source page itself names the model and says it has been released or officially announced. Rumours, leaks and planned releases are left out. The date shown is the release date given by the source, or the day we first saw the story.

Spotted a mistake or a missing model? Email hello@aidaynews.com and we will fix it.

You are free to reuse this list, including in your own articles, apps and datasets, under a CC BY 4.0 licence. All we ask is that you name AIDayNews and link back to this page. The list is also available as data (JSON), where every entry includes its own link back.

How to credit

Source: AIDayNews AI Model Release Tracker