Estimate the all-text baseline
Recognize messages, roles, tool calls, and tool results from common structured exports, then estimate the original input-token load.
Estimate what Moxby could save on a real conversation.
Upload a past conversation. See what a 5× Context Rasterization reduction could mean for context headroom and API-equivalent value.
The calculator estimates the conversation's all-text token size, models rasterized context at one-fifth of that size, and compares the difference. Actual savings vary by conversation shape, active instructions, model, caching, tools, and media. Moxby Chat Usage Reporting remains the source of truth for measured savings.
See where the context went and what comes back.
A structured 146-message coding conversation with 38 tool results.
01Import quality: Structured estimate
02All-text baseline: 184,200 estimated tokens
03Estimated rasterized context: 36,840 tokens
04Estimated context recovered: 147,360 tokens
05Estimated reduction: 80%, with about 5× the context capacity
How it works.
Apply the 5× planning ratio
Model the imported context at one-fifth of its all-text size, similar to reducing 20,000 text tokens to about 4,000 rasterized tokens.
Compare the result
See estimated context recovered, context-window headroom, and API-equivalent input value before checking measured results inside Moxby Chat.
Moxby is built for long-running work where files, decisions, tool output, and prior agent activity need to remain available with less token weight.
Other free Moxby tools.
Questions, answered.
These tools give value before installation and connect the finished plan to the Moxby browser capability built to run it.
Is the token count exact?
No. This is a 5× planning estimate from an exported transcript, not a replay of Moxby's measured app runtime. Use Moxby Chat Usage Reporting for actual savings.
What does the 5× estimate mean?
The calculator models the imported context at 20% of its all-text token size. A 20,000-token input would be represented as approximately 4,000 tokens.
Does this guarantee more subscription usage?
No. Subscription limits can include rolling windows, compute, and other policies. The report shows context-window headroom and API-equivalent value, not a promise from another provider.
What is Context Rasterization?
It represents eligible bulky, stable context as deterministic page images while keeping active instructions and exact-editing content as text. The original conversation remains stored as text in Moxby.
Four connected ways to work in your browser.
Chat on any page, build Mods, run Missions, and add ready-made products from the Marketplace.