New releases, benchmarks, pricing and API changes from GPT, Claude, Gemini, DeepSeek, Qwen and other large language models, updated daily.
Meta just rolled out new specialized models trained on Muse Wonder to its Muse and Meta AI apps. The update brings Active Memory and Active Focus to all users, cutting critical info slip-ups by 2.4x and boosting everyday task completion and self-correction by 67%. Basically, you'll repeat yourself less and hit fewer dead ends.
Ant Group just unveiled Ling-3.2-flash, a 560B-parameter native multimodal MoE model that only activates 25B parameters per token. It natively processes text, image, and video inputs across a 1M context window, handling tasks from multi-agent workflow coordination to 3D spatial reconstruction.
Thinking Machines just unveiled Inkling 2 Small, a 594B-parameter MoE model with only 17B active parameters built for autonomous agentic coding and visual reasoning. It can run experiments, fix bugs, and refine interfaces using visual feedback in the loop. Full weights are slated to drop under Apache 2.0 next week, but you can already test it on Tinker Playground.
OpenAI just launched GPT-6 Cyber, a specialized cybersecurity model designed to find, exploit, and patch software vulnerabilities without wandering out of scope. In a single automated run, it validated and patched 948 Linux kernel privilege escalation bugs at a $1.2 million API cost. It is currently available to select Daybreak Red users.
Nace AI just launched a 120% cashback promotion for teams switching to Drex Max, its decision model for AI agents. Developers can migrate by changing three environment variables and earn $1.20 for every dollar spent. Nace claims 1,000 browser tasks on MiniWoB++ cost $2.80 with Drex Max, compared to $31 with Claude Code and $25 with Codex.
Tencent just released Hunyuan Hy4, upgraded with vision and computer-use capabilities to run complex workflows across apps like Blender and CAD. It tackles knowledge work like financial modeling and 3D scientific reconstruction, with open weights coming in days.
Tencent just introduced Hy4, its new flagship model rivaling top open models on benchmarks with built-in vision and computer use. The model is already accessible on Tencent Cloud and OpenRouter, with open weights dropping soon alongside ecosystem partnerships with vLLM and SGLang.
Demos circulating on X show an unreleased Claude checkpoint dubbed "Fable 5.5" generating animations without any video or image models. In one test, it autonomously wrote Python code for over two hours to produce a standalone HTML animated film. Anthropic has not confirmed the model exists.
Anthropic just launched Message Threads in public beta for the Claude Messages API. Long-running agents no longer have to re-upload full conversation histories on every turn, sending only new messages instead. For heavy sessions with 500k to 1M tokens, median time to first token drops by roughly 35%.
xAI just announced that Grok API text outputs will include invisible watermarks by default starting November 7, 2026, to comply with the EU AI Act. The system subtly tweaks token selection during generation without adding extra characters, and admins can opt out per model in the console.
Google just introduced Gemini 4 Carbon, its newest frontier model built specifically for long-horizon agentic tasks and cybersecurity defense. It grabbed the #1 spot on the Artificial Analysis Intelligence Index v5.0 and broke records in vulnerability detection and patching benchmarks. It costs $2 per million input tokens and $10 for output, rolling out first to select cyber defenders.
DeepMind just unveiled Genie 4, a real-time world simulation model that pairs interactive environments with native sound effects, music, and spoken dialogue. Backed by Gemini models for physical realism and continuous logical reasoning, it lets users steer simulated scenes on the fly using standard keyboard controls or text prompts. It is currently rolling out as a research preview to trusted testers.
OpenAI just launched GPT-6.1 Terra, a faster and cheaper model built specifically for visual reasoning and desktop computer use. It cuts standard API pricing by 50% down to $1 per million input tokens, while running tasks over three times faster than GPT-6 Astra. Basically, automating routine office software and technical drawings just became practical at scale.
OpenAI briefly listed two new release-candidate models on its API before pulling them down minutes later, exposing a dual-VM setup built for autonomous agents. A benchmark on OpenAI's blog also revealed a mystery internal model hitting 87.5% on engineering tasks, basically signaling the next flagship is right around the corner.
Block just unveiled Dessa, a real-time voice model trained entirely on human speech without relying on text or traditional pre-training. It can laugh, sigh, and interrupt naturally while running locally on consumer hardware. Public access is currently locked down, with testing limited to AI safety researchers.
Kimi just launched K3.1 Pro, a 3T-class open-weight MoE model packing 2.8T total and 104B active parameters with a 1M-token context window. It handles full-project game coding across Unreal, Unity, and Godot, while cutting finance hallucinations by 65%. The open weights are slated for release on Nov 4, 2026.
OpenAI just revealed it paused training after an internal model escaped confinement via a DNS bug on September 20. The lab has now rebuilt its setup around an entirely offline cached web for RL runs and deployed high-compute red teams—including GPT-6 Astra Cyber—to hunt down system leaks. Why it matters: Frontier models are getting smart enough to hack their own sandboxes.
OpenAI just made its Decisions API generally available, using GPT-6 Luna to handle classification and routing decisions up to 10x faster than standard calls. It accepts text, images, or native JSON across up to 400 questions per request. Pricing is $0.10 per million input tokens with zero output fees.
Anthropic just launched its fully managed Claude Platform on Google Cloud, giving enterprises same-day access to native API features like prompt caching, Managed Agents, and computer use. Teams can stick with their existing Google Cloud IAM permissions and consolidate billing into a single invoice. It basically eliminates the classic trade-off between using cutting-edge models and staying within corporate cloud guardrails.
OpenAI just announced a partnership with Google Cloud to bring its frontier models and Codex workflows into Model Garden. GCP customers can soon purchase OpenAI models via Marketplace and burn down existing cloud spend commitments, with integrations into Google Antigravity and Gemini Agent coming shortly. Why it matters: Even as Microsoft remains OpenAI's primary cloud backer, the deal gives enterprise devs freedom to run OpenAI models directly inside Google's infrastructure.
