AI Updateshttps://ai-updates-2j9.pages.dev/What launched, what changed, what is next — checked and dated.enSat, 25 Oct 2026 09:00:00 +0500 Kimi K3.1 Pro — Moonshot's open-weight flagship, service first, weights to followhttps://ai-updates-2j9.pages.dev/models/kimi-k3-1-pro/https://ai-updates-2j9.pages.dev/models/kimi-k3-1-pro/Sat, 24 Oct 2026 09:00:00 +0500The most recent flagship-class release tracked in October: general, coding and multimodal work from Moonshot AI. Carbon-X — Hugging Face open model that chats with DNAhttps://ai-updates-2j9.pages.dev/models/carbon-x/https://ai-updates-2j9.pages.dev/models/carbon-x/Thu, 22 Oct 2026 09:00:00 +0500An open 14.8-billion-parameter model that takes raw DNA in a chat message and answers in English — for researchers, not for medical decisions. Claude Duet — proactive agent app, waitlist preview (not yet general)https://ai-updates-2j9.pages.dev/models/claude-duet-waitlist/https://ai-updates-2j9.pages.dev/models/claude-duet-waitlist/Wed, 21 Oct 2026 09:00:00 +0500Anthropic opened a waitlist preview for Claude Duet, an agent app built on Claude Opus 5.5 that works while you are away, for US Pro and Max subscribers. FLUX 3 Max — video model, price and access not yet publishedhttps://ai-updates-2j9.pages.dev/models/flux-3-max/https://ai-updates-2j9.pages.dev/models/flux-3-max/Mon, 19 Oct 2026 09:00:00 +0500A video-generation release in the same 19 Oct cluster as MiniMax's flash model. MiniMax M3.1 Flash — fast tier, price not yet publishedhttps://ai-updates-2j9.pages.dev/models/minimax-m3-1-flash/https://ai-updates-2j9.pages.dev/models/minimax-m3-1-flash/Mon, 19 Oct 2026 09:00:00 +0500A speed-focused release: general use and coding, positioned as the fast tier in MiniMax's line. Muse Spark Open — Meta's open-weight general model, licence to confirmhttps://ai-updates-2j9.pages.dev/models/muse-spark-open/https://ai-updates-2j9.pages.dev/models/muse-spark-open/Thu, 15 Oct 2026 09:00:00 +0500Meta's open release on 15 Oct, for general, coding and multimodal use. FLUX 3 Image Klein — open image model for ordinary graphics cardshttps://ai-updates-2j9.pages.dev/models/flux-3-image-klein/https://ai-updates-2j9.pages.dev/models/flux-3-image-klein/Thu, 15 Oct 2026 09:00:00 +0500The image sibling of FLUX 3 Max, released the same week, in open form. NVIDIA Nemotron 3.5 Super VL — open model for specialised work, claims to checkhttps://ai-updates-2j9.pages.dev/models/nemotron-3-5/https://ai-updates-2j9.pages.dev/models/nemotron-3-5/Thu, 15 Oct 2026 09:00:00 +0500NVIDIA's open Nemotron 3.5 Super VL, reported as beating Claude Opus 5 and GPT-6.1 Sol on specialised task benchmarks in VentureBeat's coverage. Gemini 4 Argon — agentic flagship for paid developers, price doubles 1 Dechttps://ai-updates-2j9.pages.dev/models/gemini-4-argon/https://ai-updates-2j9.pages.dev/models/gemini-4-argon/Tue, 13 Oct 2026 09:00:00 +0500Google opened Gemini 4 Argon to paid Gemini API developers on 13 Oct, after a 30 Sep defender preview. Muse Wonder and Wonder Full — weaker model by API, stronger by subscription onlyhttps://ai-updates-2j9.pages.dev/models/muse-wonder/https://ai-updates-2j9.pages.dev/models/muse-wonder/Mon, 12 Oct 2026 09:00:00 +0500Meta is directing users from Meta AI toward Muse Wonder. Only the weaker model is available via API; the strongest is subscription-only so far. GPT-6 in ChatGPT — Sol for paid, Luna for free, Astra for hard reasoninghttps://ai-updates-2j9.pages.dev/models/gpt-6-family/https://ai-updates-2j9.pages.dev/models/gpt-6-family/Wed, 07 Oct 2026 09:00:00 +0500OpenAI's GPT-6 rollout in ChatGPT from 7 Oct 2026: Sol for paid users, Luna for free users (later Help pages cast free access into doubt — check your account), Astra for complex reasoning and coding. Claude Haiku 5.5 — fastest small model, priced to match GPT-6 Lunahttps://ai-updates-2j9.pages.dev/models/claude-haiku-5-5/https://ai-updates-2j9.pages.dev/models/claude-haiku-5-5/Wed, 07 Oct 2026 09:00:00 +0500Anthropic's fastest small model, priced to match OpenAI's cheapest tier, for summarising, classifying and database work. MAI-Code-1.1-Flash — fast coding model, caps and price not publishedhttps://ai-updates-2j9.pages.dev/models/mai-code-1-1-flash/https://ai-updates-2j9.pages.dev/models/mai-code-1-1-flash/Wed, 07 Oct 2026 09:00:00 +0500A fast coding model from Microsoft AI, released the same day as Haiku 5.5. Mistral Large 4 — public preview now, weights promised by end of Octoberhttps://ai-updates-2j9.pages.dev/models/mistral-large-4/https://ai-updates-2j9.pages.dev/models/mistral-large-4/Tue, 06 Oct 2026 09:00:00 +0500Mistral's large general model, with multimodal and reasoning coverage, released 6 Oct. Nano Banana 2.1 — image model, changes not yet published herehttps://ai-updates-2j9.pages.dev/models/nano-banana-2-1/https://ai-updates-2j9.pages.dev/models/nano-banana-2-1/Tue, 06 Oct 2026 09:00:00 +0500Google DeepMind's image/multimodal release on 6 Oct, distinct from Gemini 4 Argon (language) a week later. EmbeddingGemma 2 — embedding model for search, not a chatbothttps://ai-updates-2j9.pages.dev/models/embeddinggemma-2/https://ai-updates-2j9.pages.dev/models/embeddinggemma-2/Tue, 06 Oct 2026 09:00:00 +0500A specialist release: embeddings for search and retrieval, not a chatbot. Beam — general and reasoning model, details not yet publishedhttps://ai-updates-2j9.pages.dev/models/beam-reflection/https://ai-updates-2j9.pages.dev/models/beam-reflection/Mon, 05 Oct 2026 09:00:00 +0500Reflection AI's general/reasoning/coding release on 5 Oct.