Turn product traffic into an accurate infrastructure bill. Enter your active users, daily request volume, and token profile — get instant daily, monthly and annual spend forecasts, per-user unit economics, and cross-model spend comparisons.
Daily
$0.892
Monthly
$26.76
Annual
$321.12
Per user / mo
$0.2676
| Model | Per request | Monthly | Annual | vs GPT-5.6 Luna | |
|---|---|---|---|---|---|
| Ministral 3 (8B) | $0.000294 | $8.82 | $105.84 | -67% | |
| Gemini 2.5 Flash-Lite | $0.000346 | $10.38 | $124.56 | -61% | |
| Mistral Small 4 | $0.000519 | $15.57 | $186.84 | -42% | |
| Codestral | $0.000888 | $26.64 | $319.68 | -0% | |
| GPT-5.6 Luna | $0.000892 | $26.76 | $321.12 | — | |
| GPT-5.4 nano | $0.000917 | $27.51 | $330.12 | +3% | |
| DeepSeek V4 Flash | $0.001284 | $38.532 | $462.384 | +44% | |
| Mistral Large 3 | $0.00148 | $44.40 | $532.80 | +66% | |
| Gemini 3.5 Flash-Lite | $0.001688 | $50.64 | $607.68 | +89% | |
| Gemini 2.5 Flash | $0.001688 | $50.64 | $607.68 | +89% | |
| Grok 4.3 | $0.00312 | $93.60 | $1,123.20 | +250% | |
| GPT-5.4 mini | $0.003345 | $100.35 | $1,204.20 | +275% | |
| DeepSeek V4 Pro | $0.003854 | $115.632 | $1,387.584 | +332% | |
| o4-mini | $0.003905 | $117.15 | $1,405.80 | +338% | |
| Claude Haiku 4.5 | $0.00396 | $118.80 | $1,425.60 | +344% | |
| GLM 5.2 | $0.004244 | $127.32 | $1,527.84 | +376% | |
| Muse Spark | $0.004625 | $138.75 | $1,665.00 | +418% | |
| Gemini 3.6 Flash | $0.00594 | $178.20 | $2,138.40 | +566% | |
| Mistral Medium 3.5 | $0.00594 | $178.20 | $2,138.40 | +566% | |
| Grok 4.5 | $0.00598 | $179.40 | $2,152.80 | +570% | |
| Grok 4.6 | $0.0061 | $183.00 | $2,196.00 | +584% | |
| Gemini 3.5 Flash | $0.00669 | $200.70 | $2,408.40 | +650% | |
| GPT-5.1 | $0.006825 | $204.75 | $2,457.00 | +665% | |
| Gemini 2.5 Pro | $0.006825 | $204.75 | $2,457.00 | +665% | |
| o3 | $0.0071 | $213.00 | $2,556.00 | +696% | |
| GPT-4.1 | $0.0071 | $213.00 | $2,556.00 | +696% | |
| Claude Sonnet 5 | $0.00792 | $237.60 | $2,851.20 | +788% | |
| GPT-5.6 Terra | $0.00892 | $267.60 | $3,211.20 | +900% | |
| Gemini 3.1 Pro | $0.00892 | $267.60 | $3,211.20 | +900% | |
| GPT-5.2 | $0.009555 | $286.65 | $3,439.80 | +971% | |
| GPT-5.4 | $0.0112 | $334.50 | $4,014.00 | +1150% | |
| Claude Sonnet 4.5 | $0.0119 | $356.40 | $4,276.80 | +1232% | |
| Kimi K3 | $0.0119 | $356.40 | $4,276.80 | +1232% | |
| Claude Opus 5 | $0.0198 | $594.00 | $7,128.00 | +2120% | |
| Claude Opus 4.5 | $0.0198 | $594.00 | $7,128.00 | +2120% | |
| GPT-5.6 Sol | $0.0223 | $669.00 | $8,028.00 | +2400% | |
| GPT-5.5 | $0.0223 | $669.00 | $8,028.00 | +2400% | |
| Claude Fable 5 | $0.0396 | $1,188.00 | $14,256.00 | +4339% |
Prompt caching is already saving you an estimated $3.24/month ($38.88/year) on repeated input.
Knowledge Base
Monthly cost = users × requests per user per day × 30 × (input tokens × input price + output tokens × output price), with cached input priced at its discounted rate when applicable. The calculator applies this to every model in the catalog so you can compare monthly bills side by side.
Pull real numbers from your provider's usage dashboard once you have traffic. Before launch, prototype your prompt and measure it with the token calculator, then multiply by expected conversation turns — retrieved context, system prompts and conversation history all count as input on every request.
It depends on the cached share of your input. At GPT-5's $0.125/M cached rate vs $1.25/M fresh, caching 70% of a 4,000-token prompt saves about 61% of total input cost. The calculator shows the exact monthly savings for your workload.