Models and efforts
SuperCosmos 1.1 in Line: /model, /effort, Flash, Max, Heavy and auto.
- Can Line CLI look at images or screenshots?No. SuperCosmos reads text only, so Line can't view images. Describe the screen in words, or share error text, logs or HTML instead.FAQ
- Do supercosmos-flash, supercosmos-max and supercosmos-heavy still work in Line CLI?Yes. Line accepts the per-effort ids and converts them to SuperCosmos with that effort, for example supercosmos-heavy becomes effort heavy.FAQ
- How big is the context window in Line CLI?SuperCosmos 1.1 has a 1M-token context window (1,048,576 tokens). Line auto-compacts at about 80% of it and reserves room for the reply.FAQ
- How do I change the effort (Flash, Max, Heavy) in Line CLI?Type /effort to open the picker, or /effort flash, /effort max or /effort heavy. Use -e for a single run from the shell.How-to
- How do I choose the model in Line CLI?Type /model and pick Auto or SuperCosmos. Your choice becomes the default for future sessions. Then use /effort to pick Flash, Max or Heavy.How-to
- How do I see which models Line CLI can use?Run line models. It lists Auto and SuperCosmos with their size, context window and each effort with a short description.How-to
- How do I show or hide the model's thinking in Line CLI?Use /config thinking on or /config thinking off. The setting is saved and applies to future sessions too.How-to
- How do I use a different model or effort for just one run?Start Line with -m and -e, for example line -m supercosmos -e heavy. It applies to that run only and doesn't change your saved default.How-to
- Line CLI shows "Model busy · retrying in Ns". What does it mean?When the model is very busy, Line waits and retries automatically, honouring the server's Retry-After. Your turn continues by itself.Troubleshooting
- What does the "auto" model setting do in Line CLI?Auto lets Line pick the SuperCosmos effort per step: Max for the first step and after failures, Flash for routine steps. It's the default.Explainer
- What's the difference between Flash, Max and Heavy?All three are SuperCosmos 1.1 with the same weights and 1M context. Flash is fastest, Max thinks before acting, Heavy thinks longest.Explainer
- Which effort should I use in Line CLI?Start with Auto. Use Flash for speed on simple tasks, Max for multi-file changes and debugging, and Heavy for the hardest problems.FAQ
Still need help?
Tried the steps above and still stuck? Email us. We usually reply within the first hour, and always within 24 hours. A full support team is with you.
Infrastructure questions, or permanent deletion of your data
infrastructure@clatterlabs.xyzFor deleting your whole account or the data of one service.