Salesforce just launched Slack Code, bringing coding agents like Claude, Devin, and GitHub Copilot directly into dedicated Slack channels. Teams can now watch agents generate code, review diffs, check live previews, and approve pull requests together without leaving the chat. It is available across all Slack tiers, though partner agent access is required.
Ornith.ai just launched Ornith-1.5, a family of open models built around a self-improvement loop where the AI proposes its own training tasks, builds scaffolds, and learns via reinforcement learning. The flagship 397B version matches Claude Opus 4.8 on key coding benchmarks, while the compact 9B variant can run on smartphones.
ItusiAI just open-sourced Magiviz, a turnkey platform that automates video production across five steps, from scriptwriting and character design to storyboarding, scene generation, and final stitching. It connects models like Veo, Kling, and Wan, and comes preloaded with Stripe billing and user management ready for deployment.
Z.ai just launched its GLM-5.3 API, custom-built for coding, defensive cybersecurity, and long-horizon agentic tasks. Best of all, it keeps the exact same pricing as GLM-5.2, meaning developers can upgrade their automated workflows with zero extra cost. It is available right now via Z.ai's official API and partner model gateways.
Replit just launched Free Mode, a new agent tier powered by OpenAI's GPT-5.6 Luna that lets subscribers build up to 30x more without burning credits on routine ideation and chat. Limits reset every five hours, removing the constant token anxiety before users step up to Power or Max Mode for complex tasks.
Block just open-sourced Berd, a canvas-based desktop environment built to coordinate specialized AI agents directly alongside your local files and repos. Users can spin up agents with distinct skills and personalities while swapping underlying models on the fly. It is a thoughtful push to move AI interactions away from disposable chat windows toward durable project building.
Cursor just launched Origin, a code hosting platform rolling out in early beta for all paid plan users. Origin handles repos, PRs, and real-time bidirectional sync with GitHub. Most importantly, AI agents live directly inside every repository—ready to answer questions about the codebase, make edits, and push branches without leaving the browser.
OpenAI Codex now lets users manually unlock a 1-million-token context window when using GPT-5.6 Sol. By tweaking three lines in the config file or passing CLI flags, developers can stop Codex from prematurely summarizing large codebases and tool outputs. Why it matters: massive projects stay intact in memory, though OpenAI warns its tighter defaults exist for cost and speed reasons.
Stripe just confirmed it's acquiring AI model routing startup OpenRouter for a reported $7.5 billion, a massive jump from its $1.3 billion valuation in May. OpenRouter connects 8 million developers to over 400 AI models. Why it matters: Beyond processing payments, Stripe is buying a front-row seat to AI token spending and developer demand.
Alibaba just released the open weights for Qwen3.8-27B, a native multimodal dense model under the Apache 2.0 license. At 27B parameters, it outperforms Qwen3.7-Plus and features a 262K native context window extendable to 1M tokens. Open weights for the Max-level Qwen3.8-2.4T-A95B are also available on Hugging Face and ModelScope.
SpaceX just officially closed its acquisition of Cursor AI, bringing the popular AI coding editor into its fold. The Cursor team is joining the SpaceX AI division to continue building Cursor while strengthening the broader Grok ecosystem, including Grok Build, Grok Bot, and the Grok API.
Z.ai just released GLM-5.3, a coding-focused model that achieves open-weights SOTA on Terminal Bench 3.0 and Agents' Last Exam purely through post-training scaling. Unexpectedly, the model more than doubled its cyber exploitation scores, uncovering 2,436 real-world vulnerabilities dating back up to 40 years. Model weights will drop in two weeks after a safety review.
Anthropic just shared an insider guide on optimizing token usage in Claude Code sessions. The core lesson is protecting your prompt cache: switching models or effort levels mid-conversation busts the cache, forcing full-price prefills. Engineers recommend running /clear between tasks, using @-mentions to skip file read calls, and running /compact before stepping away from the keyboard.
Claude Code desktop just added an auto-continue feature that resumes your coding tasks the moment your usage limit resets. Instead of baby-sitting long tasks and manually hitting restart, developers can now toggle a simple checkbox to let the app pick up right where it got cut off.
Google just launched Gemini 3.7 Flash, its latest workhorse model targeting software engineering and multi-step agents. It delivers major benchmark jumps—like DeepSWE climbing from 49% to 65.3%—at half the token cost ($0.75 per million input tokens). Google AI Pro and Ultra subscribers also get the new model in their Gemini Spark agent starting today.
Inherent Labs just unveiled Faraday, a 27B agent designed to replicate academic papers. Instead of writing code itself, it decouples scientific hypothesis from implementation by directing GPT-5.5 Codex as its tool, outperforming Claude Opus 4.8 and GPT-5.5. It shows that small models steering frontier tools can outmaneuver massive systems in scientific research.
DeepSeek just open-sourced DeepSeek Harness (dsh), an agent framework built entirely on an "everything-is-a-plugin" architecture. Powered by Cordis, it turns models, tools, and sandboxes into modular extensions and spins up an out-of-the-box local Web UI. It is currently in active developer preview, so expect breaking changes ahead.
xAI just rolled out Grok 4.6, a major upgrade tailored for long-running AI agents and visual development. The model matches GPT-5.6 Sol on the Artificial Analysis composite benchmark, excelling at sustained multi-step coding and self-verification. It is live now on Cursor, Grok Build, and via API starting at $2 per million input tokens.
Microsoft just rolled out MAI-Code-1.1-Flash inside GitHub Copilot, slashing inference costs by 75% while boosting token speed by 25%. The model specifically targets CLI and .NET workflows, showing a 22% jump on Terminal-Bench. A newly added 3-bit quantized version also lets developers run the 256K-context model entirely on local hardware.
OpenAI just launched a preview of the ChatGPT desktop app for Linux, supporting Ubuntu, Debian, and Fedora on both x64 and ARM64 architectures. The app lets developers hook ChatGPT, ChatGPT Work, and Codex directly into their local development workflows via .deb and .rpm packages. Why it matters: Developers finally get native, friction-free access to ChatGPT right alongside their terminal and local projects.
