Dropstone lineup history
Every Dropstone model lineup from 1.5 to 1.8, with the open-weight model that filled each tier. 1.8 is in service, 1.7 is still selectable, and 1.5 and 1.6 are retired.
Each lineup is a version, and each version is a set of three picks. This is the record of every lineup Dropstone has shipped.
| Lineup | Status | Fast | Pro | Heavy |
|---|---|---|---|---|
| 1.8 | In service | DeepSeek V4.1 Flash | GLM-5.3 Flash | Kimi K3 |
| 1.7 | Previous | DeepSeek V4 Flash 0731 | GLM-5.2 | Kimi K3 |
| 1.6 | Retired | DeepSeek V4 Flash | Kimi K2.7 Code | GLM-5.2 |
| 1.5 | Retired | DeepSeek V4 Flash | DeepSeek V4 Pro | Kimi K2.6 |
1.8 is the lineup in service. It shipped on September 6, 2026, and it is what Fast, Pro, and Heavy run today.
The Pro slot has been filled by four different models from three different makers in four lineups, and that is the point of the exercise. A tier is not a commitment to one lab.
What moved, and what that shows
| What it shows | In the lineups |
|---|---|
| A model can move down a tier | GLM-5.2 was the Heavy pick in 1.6. When Kimi K3 took Heavy in 1.7, GLM-5.2 did not leave, it moved down to Pro. A tier is a position, and models are promoted and demoted into it |
| A model can hold a tier | Kimi K3 took Heavy in 1.7 and kept it in 1.8. Re-evaluation is not a reshuffle for its own sake |
| A tier can get cheaper when the model changes | Pro moved from GLM-5.2 to GLM-5.3 Flash in 1.8, and a Pro turn now costs about what a Fast turn costs, so the same weekly allowance goes roughly nine times further on Pro |
| A tier can change model and stay where it is | Fast moved to a newer model in 1.8 and stayed included on every plan, because being on every plan is what Fast is for |
Which lineups you can still pick
The selector keeps the current lineup at the top and the previous one behind More models, so you can compare against the last generation or go back to it. In Dropstone Chat that is the button at the foot of the composer. In the CLI it is /models.
Only the immediately previous lineup is kept. That means 1.7 is still selectable and 1.5 and 1.6 are not, and a version pin works only for a lineup that is still kept. Model versions and the monthly lineup covers pinning and the numbering.
Where the write-up for a lineup lives
This page is the table. The reasoning behind each change, and what else shipped alongside it, is in the release notes. How Dropstone chooses its models covers the cycle that produces a lineup.
Related articles
- Which AI models does Dropstone use?Dropstone runs three tiers. On the 1.8 lineup, Fast runs DeepSeek V4.1 Flash, Pro runs GLM-5.3 Flash, and Heavy runs Kimi K3, re-evaluated every month.
- Model versions and the monthly lineupDropstone numbers each lineup with the year and the month it was cut, and a new version ships only when a tier's pick changes. How to read the number, and which lineup you get.
- How Dropstone chooses its modelsEvery month Blankline re-evaluates the open-weight models available for each Dropstone tier against three tests, how well it does the work, what it costs to serve, and how safely it behaves with tools. A model has to beat the one already in the slot.
- Choose a modelHow to pick between Dropstone Fast, Pro, and Heavy, where the selector lives in chat and in the CLI, and why the name Fast describes what a tier costs rather than how fast the model inside it runs.
- Which model am I using?The button at the foot of the composer names the tier, the lineup version and the effort level, and your usage page records which model handled each request.