Model releases involve the public availability of new or materially updated artificial intelligence systems from research laboratories and technology companies. Truncated covers these deployments, detailing changes in capabilities, accessibility, architecture and developer documentation.
AI-generated summary Thinking Machines Lab released Inkling 2 Small, a model with 594 billion parameters that will have its weights released next week. Artificial Analysis stated that the model scored a 31 on the Artificial Analysis Intelligence Index and noted its pricing is somewhat expensive.
AI-generated summary Google introduced Gemini 4 Carbon and Google DeepMind previewed Genie 4, while Mistral published the weights for its Large 4 model. Anthropic filed its IPO paperwork publicly and OpenAI secured about $11 billion in funding for its latest round.
AI-generated summary Salesforce introduced Koa Decisions on November 6, 2026, as a decision model built from Google’s open-weight Gemma that accepts text, images, and audio. Salesforce plans to expand the technology beyond Agentforce request routing into Salesforce workflows and developer tools.
AI-generated summary Wizstar has announced the release of Wizshot 1.0, its first in-house reference-to-video model designed to balance high visual quality with lower generation costs. The new model supports multiple reference images, generates videos up to fifteen seconds long, and is now available to logged-in users.
AI-generated summary Thinking Machines has unveiled Inkling 2 Small as its second model generation, featuring 17 billion active parameters out of a total 594 billion. Thinking Machines says the model is strongest in text recognition, infographic interpretation, and visual reasoning.
AI-generated summary A teaser titled The becoming begins has been posted on the Qwen X account. The source expects new models including Qwen4-Max, Qwen4-Flash, Qwen4-Plus, and Qwen4-27B to arrive soon.
AI-generated summary Thinking Machines released Inkling 2 Small, a 594B-total and 17B-active multimodal MoE trained in 68 hours on 12T tokens. The model features a 1M context window and is available on Tinker with Apache 2.0 weights promised next week.
AI-generated summary Google released Gemini Omni 1.2 Flash generally on AI Studio and the Gemini API featuring lower pricing and video support. Anthropic introduced Claude Personal Finance in beta for US Pro, Max, and Ultra users featuring read-only bank connections.
AI-generated summary Moonshot released the weights for Kimi K3.1 Pro, which outperformed OpenAI's GPT-6 Astra on a financial research benchmark according to the report. Nvidia says its Nemotron 3.5 Transcribe converts call recordings into text in 40 languages using a single GPU for banks.
AI-generated summary The support site release notes detail new security and admin features for Claude Enterprise plans alongside the Personal Finance beta rollout for US users. The release notes also announce new models including Claude Haiku 5.5, Claude Sonnet 5.5, Claude Opus 5.5, Claude Fable 5.1, and Claude Mythos 5.1.
AI-generated summary Huggingface.co introduces Dharma Field Extractor, a 2-billion-parameter model designed to output structured JSON fields from any document type. Huggingface.co says the model outperforms every tested frontier configuration while running on a single L40S at a lower cost per extraction.
AI-generated summary Crouton Alpha is identified as a free stealth model available on OpenRouter and OpenCode. The source reports that the model is created by Meta and shares network connections with meta/muse-spark-1.3.
AI-generated summary OpenAI launched its next-generation image generation model called Images 2.6 on Wednesday, bringing updates that include quicker use of reference images and the ability to create GIFs from text. OpenAI says that image creation is up to 50% faster than before, and the model is now available in ChatGPT, Codex, and the API.
AI-generated summary Google states that Gemini Omni 1.2 Flash is rolling out starting today to all Google AI subscribers. Google states that Google AI Ultra subscribers can now create and edit videos up to 15 seconds long.
AI-generated summary Nvidia has developed a portfolio of seven open AI models that are utilized across various industries for applications including agentic services, vision reasoning, and robotics. According to the publication, these models help companies accelerate workflows such as Dassault Systèmes using Nemotron for computer-aided engineering design and Siemens using Cosmos for factory agents.
AI-generated summary A user prompt asked Sonnet 5.5 Extra about karpathy going somewhere in 2026 without using the web. The response stated that karpathy joined anthropic in may 2026 on the pretraining team.
AI-generated summary Baidu has released the ERNIE 5.5 Flash Base open-source foundation model featuring a mixture-of-experts architecture with 280 billion parameters. According to Baidu, the model competes with similarly sized open foundation models in reasoning, code, and long-context tasks.
AI-generated summary OpenAI has debuted a new artificial intelligence model called GPT-6 Cyber that is designed to help cybersecurity professionals with tasks such as vulnerability research. OpenAI says the model helped find 81 unknown Firefox vulnerabilities and develop working exploits for hundreds of Linux kernel vulnerabilities.
AI-generated summary OpenAI says it has introduced GPT-6 Cyber, a cybersecurity model trained to find and validate exploitable vulnerabilities while staying within set scopes. OpenAI says the model scored 48.5 percent on ExploitGym and outperformed GPT-5.6 Cyber across key benchmarks.
AI-generated summary OpenAI says its security researchers built an autonomous pipeline that generated patches for 948 Linux kernel privilege-escalation bugs. Moonshot released weights for its Kimi K3.1 Pro model, which Modal describes as a post-training update with expanded RL for coding.
The top AI news in your email delivered each morning. Choose categories