The radar watched 63 sources across 32 tools this window (Sep 7 – Sep 13) and published 25 changes; these are the thirteen worth your attention, ranked. Two of them, Gemini 3.7 Flash's GA and a new anthropic-version header requirement on Claude's admin APIs, sat on hold from earlier reviews and only cleared verification this week. The rest reflects a dense week on its own, including a Cursor feature built for multi-month agent work, a paired Copilot admin control and Copilot deprecation, and DeepSeek walking back a model retirement it had already announced.

Cursor — Launch of Projects feature for managing large bodies of work with coordinator agents Cursor's new Projects feature lets you hand off work that spans months, such as a full feature or a migration, to a coordinator agent that delegates tasks across thousands of subagents and keeps working without repeated prompting. It is a step beyond single-session coding agents toward something closer to an ongoing engineering process.

Gemini API — Gemini 3.7 Flash released as general availability with improved software engineering and agentic capabilities Gemini 3.7 Flash is now confirmed generally available, positioned as Google's workhorse coding and agent model with broad improvements over the previous Flash generation. Introductory pricing runs through December 31, 2026, worth benchmarking against whatever you use today before that window closes.

GLM API — GLM-5.3-Flash pricing discount ended, prices reverted to full list prices GLM-5.3-Flash's launch discount ended on September 9; prices are back to the full list rate of $0.15 / $0.03 / $0.50 per million tokens for input, cached input, and output. Anyone who sized a workload around the discounted rate is now paying double.

Mem0 — Mem0 Pi Agent Plugin v0.3.0: Removed Dream consolidation and pin commands Mem0's Pi Agent Plugin v0.3.0 removes the Dream consolidation and pin commands along with their skills, config, types, and exports; only remember, search, forget, tour, scope, and status remain. If your integration still calls the removed commands, it breaks on upgrade rather than degrading gracefully.

GitHub Copilot — MAI-Code-1-Flash model deprecated across all GitHub Copilot experiences GitHub Copilot deprecated MAI-Code-1-Flash across Chat, inline edits, agent mode, and code completions on September 10; the suggested replacement is MAI-Code-1.1-Flash. Enterprise admins may need to enable the new model explicitly through model policies before anyone on the team can switch.

GitHub Copilot — Enterprise managed permissions for Copilot agent operations GitHub Copilot Business and Enterprise admins can now centrally control which agent operations (shell commands, file reads and edits, network domains) are blocked, require approval, or run freely, consistently across the Copilot app, CLI, and VS Code. That closes a gap where this kind of control depended on what each developer configured locally.

Anthropic API — Claude Managed Agents permission policies now support 'auto' evaluation mode Claude Managed Agents' permission policies gained an auto mode: the server itself evaluates each agent or MCP tool call and runs, denies, or pauses it for approval, reporting how the call was evaluated in the agent.tool_use and agent.mcp_tool_use events. That moves permission enforcement off the client and into infrastructure you don't have to run yourself.

Anthropic API — Claude Enterprise Admin API, Analytics API, and Compliance API now require anthropic-version header on all requests Claude's Enterprise Admin API, Analytics API, and Compliance API now require the anthropic-version header on every request, matching the rest of the Claude API. Requests without it will start failing, so check any admin tooling or scripts that call these endpoints directly before they break unannounced.

DeepSeek API — Model naming change: deepseek-v4-flash renamed to deepseek-flash DeepSeek's model naming settled on deepseek-flash (legacy deepseek-v4-flash and deepseek-v4-flash-vision-exp names still work, mapped to the same DeepSeek-V4.1-Flash backend), which now also supports a 1M token context window. Separately, DeepSeek walked back its plan to retire deepseek-v4-pro on September 14: after user pushback, it will keep serving V4 Pro at unchanged pricing instead.

OpenAI API — Prompt Cache Diagnostics now generally available in Responses API for GPT-5.6+ OpenAI's Prompt Cache Diagnostics is now generally available in the Responses API for GPT-5.6 and later: it compares cache reuse against a previous response and points out why a cache miss happened. If your prompt costs depend on cache hits, this turns a guessing game into something you can actually debug.

Together AI — Introduced preemptible compute option offering 50% discount on GPU cluster rates Together AI's GPU Clusters now offer preemptible compute, the same hardware at a flat 50% of the on-demand rate, with a five-minute drain window before a job gets interrupted. That's a meaningful discount for training or batch jobs that can tolerate a pause, less so for anything that needs to run uninterrupted.

Neon — New 'Claimable Neon' feature for agent-based development workflows Neon's new Claimable Neon lets an AI agent provision a database mid build and hand the claim link to a human afterward, instead of forcing the agent through a signup and email flow it can't complete on its own. It's a small piece of infrastructure aimed squarely at agent-driven development, not humans clicking through a console.

Claude Code — Added maxEffortLevel setting to cap effort level across all providers (Bedrock, Vertex, Foundry) Claude Code added a maxEffortLevel setting, set globally or per model under modelSettings, that caps how much effort every provider (including Bedrock, Vertex, and Foundry) is allowed to spend, while individual users can still pick something lower. For teams managing cost across a mixed-provider setup, that's a ceiling you control centrally instead of trusting everyone to dial it down themselves.

Read on the web: https://devtoolsradar.com/weekly/2026-09-13-issue-9/ (中文版: https://devtoolsradar.com/zh/weekly/2026-09-13-issue-9/)