Gemini Token Cost
Paste text to estimate tokens and the API cost on Google Gemini models. A quick heuristic, not an exact tokenizer.
Formula last reviewed 4 August 2026 · How we verify our calculators
Estimated input tokens
- Estimated cost (USD)
- $0.00241
- Input cost
- $0.00001
- Output cost
- $0.00240
Updates live as you type
Frequently asked questions
It is a heuristic blending character and word counts. Gemini’s real tokenizer (and its handling of non-text inputs) differs, so treat this as an approximation, usually within ~20%.
Flash models are optimised for speed and high volume at a low per-token rate, making them well suited to summarisation, classification and chat at scale.
No. Multimodal inputs are often priced differently from text. This tool estimates text-token cost only; check the provider’s page for media pricing.
Use Flash for simple, high-volume tasks where speed and cost matter; use Pro for harder reasoning and quality-critical work where the higher rate is justified.
Sixteen tokens of instruction, 800 tokens of summary — guess which one costs more
"Summarise the key points of this quarterly report in five bullets." estimates to roughly 16 input tokens under this calculator's heuristic. Paired with 800 expected output tokens on Gemini Flash, output cost dominates the total — the same pattern seen on the ChatGPT and Claude estimators, since a short instruction generating a longer reply is a fundamentally output-heavy request no matter which provider's model handles it.
What the estimate can and can't be trusted for
Token counts here come from a heuristic blending character and word counts, not Gemini's actual tokenizer, which remains the only exact source. For ordinary English prose the estimate is usually within about 20%, though text heavy with code, numbers or non-English characters can tokenize quite differently from plain prose — worth keeping in mind before trusting this closely for technical or multilingual prompts. Images and audio are typically priced on a separate schedule entirely; this tool covers text tokens only.
Why Flash exists, and when Pro is worth the extra cost
Gemini's line-up spans a deliberate cost-capability trade-off: Flash is optimized for speed and high volume at a low per-token rate, well suited to summarisation, classification and chat at scale, while Pro targets harder reasoning at a meaningfully higher rate. Pricing runs per million tokens, output priced above input on both tiers.
Why this category is worth re-checking often
Gemini competes aggressively on price for high-volume Flash usage specifically, and per-token rates in this space have historically dropped release over release, sometimes substantially within a single year — worth re-running this calculator whenever a new Flash generation ships rather than assuming last year's numbers still hold.
Pricing as of June 2026. LLM rates change frequently — verify current prices on the provider's official pricing page before budgeting.