Ai Releases
The latest AI Releases from the internet.

ChatGPT Wallet Explained: How OpenAI's Agent Payments Work
OpenAI's new wallet lets AI agents complete purchases for you without ever receiving your real card details.
Read the featureAi Releases
-
Read →DeepSeek API App Features: Search, Files and Voice for Developers
DeepSeek has opened the capabilities behind its consumer app to developers, who can now add web search, file input and voice to their own products.
-
Read →Antigravity Orchestrator Pairs Gemini 4 Carbon and Hydrogen
Google says Orchestrator, a new Antigravity model option, runs Gemini 4 Carbon for planning and Hydrogen for execution in one session, starting with…
-
Read →ChatGPT Dot Custom Rules Set When a Dot Can Act
ChatGPT says its dots can now follow custom instructions, and its controls docs describe four rules for when a dot may take an action.
-
Read →MiniMax H3-Image Opens Weights for 4K Image Generation
MiniMax says its H3-Image model has open weights on Hugging Face, and its API docs list 4K output, editing and per-image pricing.
-
Read →Claude Fable 5.2 and Mythos 5.2 Launch With Cheaper Fable
Anthropic has released Claude Fable 5.2 and Mythos 5.2, cut Fable's API price to $8 and $40 per million tokens, and limited Mythos to vetted…
-
Read →LongCat 2.5 Is Free on OpenCode for Two Weeks
OpenCode says LongCat 2.5 is free on its platform for two weeks, with a 1M-token context, multimodal input and zero data retention.
-
Read →GPT-6 Cyber Launches for Select Daybreak Red Customers
OpenAI launches GPT-6 Cyber for select Daybreak Red customers, with access expanding over time.
-
Read →SWE-2 Flash Is a Fast Coding Model Free Through November
Cognition's new SWE-2 Flash coding model is free through November 30 for paid Devin users in Devin Desktop and Devin CLI, with its weights held back…
-
Read →Tencent Hy4 Gains Vision and Computer Use at Preview Prices
Tencent's Hy4 model is out of preview with native vision and computer use, keeps its preview API prices, and has open weights and a Flash variant…
-
Read →ChatGPT Business 1 Seat Option Limited to Some New Workspaces
OpenAI's Help Center says some eligible customers can open a ChatGPT Business workspace with one paid seat, though its own pages disagree on the…
-
Read →Meituan's LongCat-2.5 Launches as a 1.6T Agentic Model
Meituan has released LongCat-2.5, a 1.6-trillion-parameter multimodal agentic model available through its chat service and API, with open weights…
-
Read →Tencent's Hyra 2.0 Builds Reusable Research Tools and Models
Tencent Hunyuan has launched Hyra 2.0, a research agent framework that builds reusable scientific models and tools, with protein, quantum and proof…
-
Read →Mistral Large 4 Open Weights Land on Hugging Face
Mistral has released open weights for Mistral Large 4, its Le Chonk model, which it prices at $0.70 per million input tokens through its API.
-
Read →Nemotron 3.5 Transcribes 40 Languages and Labels Speakers in 13
NVIDIA has released Nemotron 3.5 Transcribe, an open-weight speech model that combines transcription, speaker labels, translation and audio…
-
Read →Claude for Legal Artifacts Gallery Offers Ready-Made Legal Tools
Anthropic has launched a Claude for Legal Artifacts gallery with ready-made tools for contract review, regulatory horizon scans and cite checks.
-
Read →MAI-Image-2.7-Preview Hits No. 3 on Arena's Image Edit Board
Microsoft AI has released MAI-Image-2.7-Preview for image editing, and MAI-Image-2.6 and 2.6-Flash are now generally available in Microsoft Foundry.
-
Read →ChatGPT iCloud Mail and Calendar Plugins Roll Out to All Plans
OpenAI has made iCloud Mail and Calendar plugins available in ChatGPT on all plans across web, desktop and iOS, with setup requiring an Apple…
-
Read →Nanbeige Says Nanbeige4.3-33B Leads Open ~30B Models on Agentic Tasks
Nanbeige has released the open-weight Nanbeige4.3-33B, claiming leading agentic coding and cowork scores among ~30B open models based on its own…
-
Read →MiniMax M3.1 Flash Ships Open Weights at a Quarter of M3's Price
MiniMax has released M3.1 Flash with open weights on Hugging Face and API pricing of $0.15 per million input tokens, a quarter of M3's list rate.…
-
Read →Canva's First Image Model Lands Behind Only OpenAI
Canva's first in-house image model, Craft, is live in Dream Lab for all users and on the public API. Canva's chart puts it 4th on DesignArena at 1340…
-
Read →Grok Bot can now build reports and dashboards that stay updated
xAI's @bot account says Grok Bot can now build editable pages that it keeps up to date. The docs, last updated before the announcement, don't…
-
Read →Ornith-1.5-122B Released: MIT-Licensed Open Weights, Quantised Builds and Ollama Support
Ornith-1.5-122B is a new open-weight reasoning MoE model under the MIT license with a 262,144-token context. This article covers how to run it with…
-
Read →Mercury 3: Haiku-Class Quality at About 4x the Speed, Per Inception's Own Numbers
Inception says its diffusion LLM Mercury 3 matches or beats Claude Haiku 5.5 and GPT-6 Luna on quality while running 4–8x faster. Here is the pricing…
-
Read →Ternary Bonsai 2 2B, 4B, 9B: Smallest Files Need PrismML Builds
PrismML's Ternary Bonsai 2 2B, 4B and 9B are out under Apache 2.0. The 0.43–2 GB sizes need PrismML's binaries; stock llama.cpp loads larger Q2_0…
-
Read →Cat Jam Free Is Free on OpenCode, No End Date Announced
OpenCode's Cat Jam Free, a text-only reasoning model with about 262K context, is free on Zen and Go. Its docs note 30-day data retention.
-
Read →Muse Wonder Is Out: Where to Use It and How It Benchmarks
Meta's Muse Wonder is available in the Meta Model API and Muse Code, and Muse Wonder Full is for Muse Code subscribers. Here is who can use each,…
-
Read →Claude Haiku 5.5 is out, on average about 75% cheaper than Haiku 4.5
Anthropic has released Claude Haiku 5.5, a low-cost small model it says costs about 75% less to run than Haiku 4.5 on average. It adds an adjustable…
-
Read →Exo Free Listed as a Limited-Time Free Model on OpenCode
Exo Free is listed in OpenCode Zen as a limited-time free model, with reported support for long context, images and tool calls, but with an important…
-
Read →Nano Banana 2.1: Reported Improvements and Availability
An X post announces Nano Banana 2.1 as an image-generation and editing update with reported quality improvements, bug fixes, lower pricing and access…
-
Read →Mistral Large 4: API Access and Open-Weights Timeline
Mistral announces Large 4, a natively multimodal model with 1 trillion total parameters, API access today and open weights planned for October.
-
Read →Banbury Road AI’s Kardashev-0.7 Is a 32-Model Swarm
Banbury Road AI describes Kardashev-0.7 as a 32-model trained swarm using RLPS to develop complementary specializations, with reported efficiency…
-
Read →Reflection AI Beam Model: 501B Parameters, 23B Active
Reflection AI has released Beam, an open-weight model with 501 billion total parameters and 23 billion active parameters, under the Apache 2.0…
-
Read →GLM-5.3 UNCENSORED EXL3 3.0bpw: What the Release Includes
A source-based explanation of the community-hosted GLM-5.3 UNCENSORED EXL3 quant, its model lineage, reported tests, TabbyAPI setup and tool-calling…
-
Read →Microsoft MAI-Transcribe-2-Streaming: Reported Specs and Price
An attributed explanation of the reported accuracy, latency, ranking and pricing figures for Microsoft’s MAI-Transcribe-2-Streaming.
-
Read →Gemini 4 Argon: announcement, leaks and reports on access and price
Google says Gemini 4 Argon is now rolling out globally in the Gemini API and to paid Gemini subscribers, after a first rollout to trusted testers.…
-
Read →Ideogram 4.5: Access, Pricing and Open Weights
Ideogram 4.5 is reported as a precise-editing image model available through Ideogram, APIs and partner services. Open weights are planned, not…
-
Read →What Are Dots? The AI Agents Announced With Astra
An announcement describes Dots as always-on AI assistants powered by Astra, with computer, browser and app access, Pro-plan inclusion and planned…
-
Read →Claude Sonnet 5.5 launches after weeks of leaks, Anthropic says
Anthropic has released Claude Sonnet 5.5 after weeks of leaks about stealth testing, Claude Code references and a registry entry. The speed, cost and…
-
Read →Manus 2.0 brings a new agent architecture and creative tools
Manus 2.0 combines a new agent architecture with Manus Studio, event-triggered Automations, Cloud Computer, Computer Use and the personal-agent app…
-
Read →Naive-N0.5-Flash: Inside NaiveAI’s 309B Open-Weight Model
Naive-N0.5-Flash is NaiveAI’s 309B open-weight MoE model for coding and AI R&D, with a native 1M-token context and the NaiveRT inference runtime.
-
Read →MiniMax M3.1-Flash-Preview arrives on MiniMax Code
MiniMax announces M3.1-Flash-Preview for MiniMax Code, along with double-credit check-ins and Token Plan quota resets.
-
Read →Julia 1 Is a Compact Model for Finite-Choice Decisions
Supersonic Labs’ Julia 1 is a 144.3-million-parameter model for classification, routing, ranking and yes-or-no decisions, with Python CPU and browser…
-
Read →Pixel Canary Is Free in Cline, Cline Says
Cline says its newly released Pixel Canary stealth model is free in Cline and reports a 90% pass@4 result on Next.js Agent Evals.
-
Read →Google introduces Gemini 3.8 Flash TTS models for expressive audio
Google has announced Gemini 3.8 Flash TTS and Flash-Lite TTS, with access through the Gemini API and Google AI Studio.
-
Read →GPT-6 Sol and Luna: From Early Leaks to Launch and GPT-6.1 Updates
OpenAI's GPT-6 Sol and Luna launch has moved from months of leaks and speculation to official API and ChatGPT availability. Later reports track…
-
Read →Claude Opus 5.5: from stealth-test reports to Anthropic's announcement
Anthropic has officially announced Claude Opus 5.5, while the earlier leaks, code observations and pricing figures remain unverified. Availability,…
-
Read →Tencent Hunyuan Launches Hy Image3.5 Preview
Tencent Hunyuan has announced a Hy Image3.5 preview with text-to-image and image-to-image generation, up to 2K output, and stated API pricing of…
-
Read →Xiaomi MiMo-V2.6 Pro and Flash: Reported Benchmarks
Xiaomi announces MiMo-V2.6 Pro and Flash with open model weights and reports mixed benchmark results against MiMo-V2.5 and named frontier models.
-
Read →Grok 4.7 timeline: from release countdown to xAI's announcement
xAI has announced Grok 4.7 and listed its prices. Earlier leaks and sightings are tracked here, with company-reported benchmark results kept separate…
-
Read →Kimi Code Desktop brings multiple coding agents into one workspace
Kimi describes Kimi Code Desktop as a workspace for managing multiple coding agents, running tasks in parallel and coordinating long-running…
-
Read →Qwen-Image-2.1: Open-Weight Image Generation and Editing
A grounded overview of Qwen-Image-2.1’s documented image-generation, editing, transparency, reference-image, and developer-integration features.
-
Read →Ternary Bonsai 2 27B Reports 98.2% of Qwen3.8’s Score at 5.9 GB
PrismML says Ternary Bonsai 2 27B is a 5.9 GB model based on Qwen3.8 27B, with 98.2% of its reported aggregate benchmark performance under Apache 2.0.
-
Read →QuiverAI Introduces Arrow 2.0 for Editable SVG Graphics
QuiverAI introduces Arrow 2.0 for generating, editing, vectorizing and animating editable SVG graphics, alongside the higher-refinement Arrow 2 Telos.
-
Read →Union Alpha is free for a week, says OpenCode
OpenCode describes Union Alpha as a stealth model for agentic coding with image support and says it will be free for one week.
-
Read →Kimi K2.8 Preview: What Changes in Kimi Code
Kimi K2.8 Preview is fully rolled out in Kimi Code with the existing model ID, adjustable thinking effort and up to 1 million tokens of context.
-
Read →Cognition’s SWE-2 Targets Better Coding Performance at Lower Cost
Cognition’s SWE-2 adds medium, high, and max effort levels, with company-reported gains in coding benchmarks and efficiency over SWE-1.7.
-
Read →ChatGPT Images 2.5 Features: Faster Generation, Sketch and More
ChatGPT Images 2.5 adds faster generation, improved fidelity, more consistent edits, comment-based changes, Sketch, and templates for formats such as…
-
Read →DeepSeek-V4.1-Flash-1017 is live
DeepSeek has released DeepSeek-V4.1-Flash-1017, which replaces the earlier beta build and is served under the deepseek-flash model name.
-
Read →GPT-6 Astra Signals a New Era of Computer-Using AI
A brief announcement introduces GPT-6 Astra with a broad claim: it can perform any computer-based task quickly. Here are the benchmark, pricing and…
No entries match that filter.
