Google Research: Introduced TabFM, a foundation model for tabular data classification and regression that can generate high‑quality predictions on unseen tables in a single forward pass. ClaudeDevs: Raised Claude Platform API rate limits for all users, removed spend‑based tiers, and gave the latest Sonnet and Haiku models five‑times higher limits at the top tier. Google AI: Launched two workflow updates - one model for ultra‑fast image generation and another to instantly animate those images, both at a fraction of the usual cost, highlighted by Nano Banana 2 Lite. GitHub Changelog: Announced that Gemini 2.5 Pro and Gemini 3 Flash will be deprecated in all Copilot experiences on July 31 2026, urging migration to Gemini 3.1 Pro or Gemini 3.5 Flash beforehand. Z.ai: Rolled out ZCode, the official development environment for GLM‑5.2, offering 1.5× usage quota for GLM Coding Plan subscribers, BYOK support, and cross‑platform availability on macOS, Windows, and Linux. GitHub: Made Kimi Moonshot’s Kimi K2.7 Code generally available in GitHub Copilot as the first open‑weight model selectable in the model picker, noting lower cost with performance comparable to top‑tier models.
Qwen: introduces Qwen3.7-Plus, a multimodal agent model that unifies vision and language with both GUI and CLI operation and serves as a coding and productivity assistant. OpenAI: frontier models and Codex are now generally available on AWS via Amazon Bedrock, extending enterprise security, compliance, and governance workflows. xAI: Composer 2.5 is now inside Grok Build, described as a fast, highly intelligent model for long‑running tasks and complex instructions. LangChain: highlights Fleet for secure agent access to private resources and adds LangSmith LLM Gateway spend limits that return a 402 error when caps are hit. Google Antigravity: is becoming a scientific workbench with a Science Skills bundle that runs complex workflows like protein analysis using Alpha* models and dozens of databases; Google Gemma: releases the first gemma‑skills iteration, enabling agents to build with Gemma, use MTP for speed, pick model size, and locate up‑to‑date resources. ClaudeDevs: resets 5‑hour and weekly rate limits for Pro/Max plans and fixes excessive parallel subagents; Cursor: raises usage limits for Teams and adds a Premium seat with 5× usage at 3× cost; Visual Studio Code: demos orchestrating agents via the VS Code Agents window; NVIDIA: adds real‑time AI media tools including Synthetic Video Detector (up to 92% accuracy, 22 ms latency), RTX Video Super Resolution and Frame Generation; Vercel: enables remote execution of Conductor’s parallel coding agents on fast Sandboxes; Perplexity: launches Search as Code, a new architecture that writes Python to call its search stack directly, now default in the Perplexity Agent API.
ClaudeDevs (@ClaudeDevs) announced an update to Opus 4.8, which now allows for the addition of system instructions mid-conversation without breaking the prompt cache. This improvement is expected to reduce the cost and latency of API requests. The update aims to enhance the efficiency of the system.
