Prices and cost
Prices per 1M tokens, long-context input, and worked cost examples.
- Am I charged for SuperCosmos's thinking?Yes. Reasoning tokens on Max and Heavy are output tokens and billed at the output price.FAQ
- Did prices change with SuperCosmos 1.1?No. 1.1 kept the prices set on Oct 31, 2026: Flash $1.60 / $3.40, Max $1.80 / $3.80, Heavy $2.00 / $4.40.FAQ
- Do I pay for my system prompt on every request?Yes. Everything you send is input, every time, including the system prompt and history. Keep it concise.FAQ
- Does a 262K-token prompt cost the same as a 263K-token one?No. Crossing 262K tokens switches the whole input to 1.5x, so a slightly longer prompt can cost noticeably more.FAQ
- How can I estimate the cost of a request before sending it?Estimate input tokens from your prompt, guess output length, and multiply by the tier's prices; check usage afterwards.How-to
- How does SuperCosmos usage count toward my plan in Line CLI?App usage is measured as what requests would cost at model rates, so Heavy uses up your limits faster than Flash.Explainer
- How far does $5 of credit go on SuperCosmos?$5 buys about 3.1M input or 1.5M output tokens on Flash, less on Max and Heavy. Real use mixes both.FAQ
- How have SuperCosmos prices changed over time?Launch prices (Oct 27, 2026) were Flash $0.75/$2.50, Max $3.00/$5.00, Heavy $6.00/$11.51; since Oct 31 they're $1.60/$3.40, $1.80/$3.80, $2.00/$4.40.Reference
- How much did 1,000 benchmark tasks cost on each tier?In Benchmark 2, 1,000 tasks cost $1.64 on Flash, $2.57 on Max and $3.64 on Heavy at list prices.Reference
- How much do 1,000 short chat messages cost?Worked example at SuperCosmos 1.1 list prices: 1,000 exchanges of about 500 input and 200 output tokens each.FAQ
- How much does a 500K-token prompt cost?Worked example at SuperCosmos 1.1 list prices: 500,000 tokens in (over 262K, so input bills at 1.5×), 2,000 tokens out.FAQ
- How much does a long, deeply reasoned answer cost?Worked example at SuperCosmos 1.1 list prices: 2,000 tokens in and 20,000 tokens out (reasoning plus answer).FAQ
- How much does it cost to ask a question about a 200K-token codebase?Worked example at SuperCosmos 1.1 list prices: 200,000 tokens of code in, a 2,000-token answer out (still under the 262K long-context threshold).FAQ
- How much does it cost to classify 100,000 short texts?Worked example at SuperCosmos 1.1 list prices: 100,000 requests of about 300 input tokens and 5 output tokens (a one-word label).FAQ
- How much does it cost to fill the whole 1M-token window?Worked example at SuperCosmos 1.1 list prices: 1,048,576 tokens in (long-context rate) and 1,000 tokens out.FAQ
- How much does it cost to summarise a long document?Worked example at SuperCosmos 1.1 list prices: A 50,000-token document in, a 1,000-token summary out.FAQ
- How much does it cost to translate a book-length text?Worked example at SuperCosmos 1.1 list prices: About 120,000 tokens in and about 130,000 tokens out, sent in chunks under 262K tokens.FAQ
- How much does one short chat message cost?Worked example at SuperCosmos 1.1 list prices: A short question (about 150 input tokens) and a short answer (about 100 output tokens).FAQ
- How much does SuperCosmos cost?SuperCosmos 1.1 per 1M input / output tokens: Flash $1.60 / $3.40, Max $1.80 / $3.80, Heavy $2.00 / $4.40.Reference
- How much would a chatbot with 10,000 daily users cost?Worked example at SuperCosmos 1.1 list prices: 10,000 users × 5 messages a day × 30 days, each message ~800 input and ~250 output tokens.FAQ
- Is Heavy worth paying more for?Only for hard problems. Heavy costs more per token and writes more tokens; on everyday tasks Flash scores about the same.FAQ
- Is SuperCosmos the same price in every product?Yes. Token prices are the same in Line CLI, the Model API, the Agent API and Cosmos Agent; how you pay differs.FAQ
- Is there a discount for repeated prompts (prompt caching)?No caching discount is published. Every request bills its full input at the tier's price.FAQ
- Was there a free launch offer for SuperCosmos 1.1?Yes. Free-plan users got SuperCosmos Flash free in Line CLI and Cosmos on Telegram until 00:30 UTC on November 3, 2026.FAQ
- What is the long-context price (1.5× above 262K tokens)?If a request's prompt is over 262K tokens, its whole input is billed at 1.5× the normal input price. Output is unchanged.Explainer
- What's the cheapest way to use SuperCosmos?Use Flash, keep prompts and answers short, trim history, stay under 262K input tokens, and use the Free plan's allowance in the apps.FAQ
- Why did my Heavy request cost much more than Flash?Heavy has a higher price per token and writes more tokens (reasoning plus longer answers), so cost per task rises faster.FAQ
- Why do output tokens cost more than input tokens?Writing tokens takes more compute than reading them, so every tier prices output higher than input.Explainer
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.