Free
$0
$0.20usage included
- Input
- $0.05 / 1M tokens
Sage / Pricing
Why Sage.
Speed10× faster than a frontier LLM
Median, vs Claude Opus 5 on the same decisions.Show the mathHide the math
Sage330 ms183 ms without network
GPT-5.6 Terra1,110 ms3.3× slower
Claude Opus 53,190 ms9.6× slower
Claude Sonnet 55,020 ms15.0× slower
CostOver 100× cheaper than a frontier LLM
Per decision, on the same questions.Show the mathHide the math
Sage — Pro$0.000017400 input tokens at Pro rates
GPT-5.6 Terra$0.0009355× more
Claude Sonnet 5$0.002over 100× more
Claude Opus 5$0.0042over 100× more
IntelligenceMore accurate than other decision models
It reasons through hard questions, which no other decision model does, and reads images.Show the mathHide the math
Sage v1.1Jev 1.13
JevBench v1.2, public — 213203vs186
multistep_decisions — 10094vs71
Plans
Developer
$14/mo
$14usage included
- Input
- $0.05 / 1M tokens
Starter
$49/mo
$49usage included
- Input
- $0.046 / 1M tokens
Pro
$99/mo
$99usage included
- Input
- $0.042 / 1M tokens
Priority support
Growth
$249/mo
$249usage included
- Input
- $0.038 / 1M tokens
128K context
Priority support
All plans: Output $10 / 1M tokens · Web search $0.10 each
Start free
No seats, no overage invoices: when included usage runs out, calls return 402 until the next period or an upgrade. Unused usage doesn't roll over. Higher volume: team@levanto.ai
On every plan
Yes / No
Choice
Scale
Sort
Tags
Batching
Reasoning
Images
Web search
How usage is counted
- Input tokens
- Everything Sage reads: your content, question, instructions and options. Images count as tokens. Content counts once per question, once per tag and once per item to sort.
- Output tokens
- Sage’s reasoning, when it thinks a question through. Answers are free.
- Web searches
- Charged per search that runs.
CallTokens (approx.)DeveloperGrowth
One question, short message400 in · 0 out$0.00002$0.000015
One question, 8,000-token document8,400 in · 0 out$0.00042$0.00032
Ten questions, same document, one batch84,000 in · 0 out$0.0042$0.0032
Five tags, short message2,000 in · 0 out$0.0001$0.000076
Every response reports input_tokens, output_tokens and image_tokens in meta.usage. POST /usage/estimate returns them before you send.
