The everyday workhorse, and the most substantive change in 1.7: last cycle's Heavy tier is now the default on every plan. Built on GLM-5.2, hosted in the US.
Pro moves to GLM-5.2, 1.6's Heavy tier, now the default on every plan including free. 51.1 on the independently measured AA Intelligence Index, fourth on the Frontend Code Arena ahead of Claude Opus 4.8. Kimi K2.7 Code leaves the user-facing cascade but stays registered as the vision backend.
Pro tracks the best open-weight model for tool-driven, day-to-day coding. The version number bumps; the CLI does not change.
Dropstone Pro is the default for the bulk of real engineering: planning, multi-file edits, running tests, and tool-driven workflows where Heavy's budget is unnecessary. At 1.7 it runs Z.ai's GLM-5.2, a 744B-param Mixture-of-Experts model (roughly 40B active per token, MIT licensed, 1M context) that was our paid-only Heavy tier six weeks ago. It scores 51.1 on the independently measured Artificial Analysis Intelligence Index, above Gemini 3.1 Pro at 46.5, and ranks fourth on the Frontend Code Arena at 1587, ahead of Claude Opus 4.8, across 1,757 blind developer votes. Moving it down a tier is not a demotion of the model; it is a statement that what counted as our top tier last cycle is now our default, available on every plan including free. Pro is text-only, so image requests route to Kimi K2.7 Code, the registered vision backend. Same approval-gated CLI, same US-hosted inference.
Plan a feature, edit across the repo, run the test suite, fix the failures. One approval gate per tool call.
GLM-5.2 ranks fourth on the Frontend Code Arena, ahead of Claude Opus 4.8, where developers pick the better output without knowing which model produced it.
Run Pro across a PR diff to flag security, performance, and style issues, with the same OWASP coverage as Dropstone's safety analyzer.
Every Dropstone request is treated as if the model could be adversarial. The CLI requires explicit user approval before any action that writes to disk, runs a shell command, or fetches a URL. No model output is ever auto-executed.
Pro runs on SOC 2-certified, US-based inference providers. Enterprise and commercial data is never stored and never trained on, regardless of any setting. Consumer sessions help improve the open model we run and can be turned off anytime in settings, as text only, never your raw images.
Pro is built on GLM-5.2, an open-weight foundation model. Goldwasser et al. (2022) proved no party can prove a closed foundation model is free of embedded behaviors, including Anthropic for Claude and OpenAI for GPT. We say this out loud. The runtime is why model origin does not matter for your code.
Same approval gate, same US-hosted inference, same zero-retention guarantee across all three. Pick the smallest model that meets the task.
Low-latency agentic coding that is no longer just a latency tier. Built on DeepSeek V4 Flash 0731, hosted in the US.
The everyday workhorse, and the most substantive change in 1.7: last cycle's Heavy tier is now the default on every plan. Built on GLM-5.2, hosted in the US.
The strongest open-weight model measured: first open model past a flagship closed model on the aggregate index, and #1 on the Frontend Code Arena. Built on Kimi K3, hosted in the US.
Fast and Pro are included on every plan, including free. Heavy is unlocked on the Pro plan ($20/mo) and Max plan (from $100/mo).
Install the CLI, authenticate, and start running approval-gated agentic workflows. No credit card to start.
