~/tools / token-meter
Token & cost meter
Paste any prompt or document and get an instant token estimate, then compare what it costs to run across six model families — per call, per day, per month. Rates are editable inline because prices drift; the context-fit bar warns you before you blow the window. Built from the cost-observability dashboard I shipped in production.
TOOL 01
Token & cost meter
born from the LLM cost dashboard I shipped at Jagriq| model | in $/MTok | out $/MTok | per call | per day | per month | context fit |
|---|---|---|---|---|---|---|
| claude-sonnet | $0.0120 | $12.00 | $360 | |||
| claude-haiku | $0.0040 | $4.00 | $120 | |||
| gpt-5 | $0.0080 | $8.00 | $240 | |||
| gpt-5-mini | $0.0016 | $1.60 | $48.00 | |||
| gemini-pro | $0.0080 | $8.00 | $240 | |||
| bedrock-llama | $0.0006 | $0.5760 | $17.28 |
Estimates ~4 chars/token. Rates are editable (dashed cells, $/MTok) — prices drift, your numbers shouldn't.
more
tools.siblings
RAG capacity plannercorpus size → vectors, RAM, embed costProvider request translatorOpenAI ⇄ Anthropic/Bedrock request JSONGPU VRAM calculatorwill that model fit on your card?JSONL dataset inspectorfine-tune data sanity check, fully localPrompt diffword-level diff for prompt iterationPrompt linterESLint for system prompts