{"@context":"https://obfus.link/schema/tool/v1","tool":{"name":"Dev Unit Converter","slug":"dev-unit-converter","mcpToolName":"dev_unit_converter","tier":"tier_1","tierLabel":"Tier 1 — Commodity","category":"converters","canonicalUrl":"https://obfus.link/tool/dev-unit-converter","endpoint":"https://obfus.link/mcp","priceMicros":8000,"tagline":"Convert data, tokens, CSS, and time units with LLM tokenizer ratios","atomicAnswer":"Dev Unit Converter handles developer-specific unit categories that generic converters miss: data with IEC/SI base toggle, CSS units with configurable viewport and font-size, LLM tokens with model-specific tokenizer ratios for GPT-4, Claude, Llama-3, and Gemini, plus time and frequency. Every conversion returns the numeric result, a formatted string, and the formula.","description":"Converts developer-specific units across five categories: data (IEC/SI), CSS (px/rem/em/vw/vh with configurable viewport), LLM tokens to words with model-specific ratios for GPT-4, Claude, Llama-3, and Gemini, time, and frequency. Returns the result, formatted string, and the conversion formula.","inputSchema":{"type":"object","required":["value","from","to","category"],"properties":{"to":{"type":"string"},"from":{"type":"string"},"value":{"type":"number"},"category":{"enum":["data","tokens","css","time","frequency"],"type":"string"},"binaryBase":{"enum":["iec","si"],"type":"string"},"tokenModel":{"enum":["gpt-4","gpt-4o","claude-sonnet","claude-opus","llama-3","gemini"],"type":"string"},"baseFontSize":{"type":"number"},"viewportWidth":{"type":"number"},"viewportHeight":{"type":"number"}}},"outputSchema":{"type":"object","properties":{"note":{"type":"string"},"result":{"type":"number"},"formula":{"type":"string"},"formatted":{"type":"string"}}},"agenticReasoning":"USE THIS WHEN: (1) You need to convert data sizes between byte units and must choose between IEC (1024-based: KiB, MiB, GiB) and SI (1000-based: KB, MB, GB) — specify binaryBase to select the base, or use IEC-named units (kib, mib, gib) which auto-select IEC. (2) You are estimating LLM API costs or prompt token budgets and need to convert between tokens, words, characters, and pages with model-specific accuracy — the token-to-word ratio differs between GPT-4 (0.75), Claude (0.68), Llama-3 (0.72), and Gemini (0.70). (3) You are generating CSS values and need to convert between px, rem, em, and viewport-relative units with the correct base font size and viewport dimensions. DO NOT USE WHEN: you need temperature, weight, distance, or currency conversion — this tool covers developer-context units only. OVER ALTERNATIVES: prefer over manual arithmetic (no formula, easy to confuse IEC vs SI) and general-purpose converters (no CSS viewport parameters, no LLM tokenizer ratios).","mcpDescription":"Converts developer-specific units across five categories. USE WHEN: converting bytes between IEC and SI bases, converting CSS px to rem with a custom base font, estimating LLM token budgets across model-specific ratios, or converting time and frequency units. INPUT: value, from, to, category, plus category-specific options. OUTPUT: result number, formatted string, conversion formula, optional note for tokens. COST: 1 unit.","howTo":[{"step":"Select category","description":"data, tokens, css, time, or frequency."},{"step":"Enter value and units","description":"Provide the numeric value and source/target units."},{"step":"Configure options","description":"For data: choose IEC (1024) or SI (1000). For tokens: pick the LLM model. For CSS: set base font size and viewport."},{"step":"Convert","description":"Click Convert or press Ctrl+Enter. The result shows the formatted value and the exact formula used."}],"faqs":[{"question":"What is the difference between KB and KiB?","answer":"KB (kilobyte) in SI units equals 1,000 bytes. KiB (kibibyte) in IEC units equals 1,024 bytes. Hard drive manufacturers use SI (1,000-based), while operating systems typically report sizes in IEC (1,024-based). Select the base using the IEC/SI toggle in the data category."},{"question":"Why do token-to-word ratios differ between models?","answer":"Different models use different tokenizers. GPT-4 and GPT-4o average roughly 0.75 words per token. Claude models average about 0.68 words per token due to a different tokenizer. Llama-3 averages 0.72 and Gemini 0.70. These are approximations — actual ratios vary by language and content type."},{"question":"How do I convert vw or vh CSS units?","answer":"Select the css category, choose vw or vh as source or target unit, and enter your viewport width (for vw) or height (for vh) in the configuration fields. The conversion requires knowing the viewport dimensions."},{"question":"Can I use this tool via the MCP API?","answer":"Yes. The tool is registered on the obfus.link MCP server at https://obfus.link/mcp. Call it from any MCP-compatible agent with a Shared Payment Token. The MCP tool name matches the snake_case slug shown in the integration snippet."}],"workflowChains":{"live":[],"planned":[]},"tags":["units","convert","tokens","css","px","rem","bytes","llm"],"tddVerified":true,"mcpCostUnits":1}}