Customer Support Bot API Cost Calculator (Budget Tier)
Calculate high-throughput customer support and ticketing AI bot costs using budget models like Gemini Flash Lite, DeepSeek V4.1 Flash, and GPT-6 Luna.
Describe Your Workload
= 150,000 requests a month (30 days)
System prompt + chat history + retrieved documents + the user's message.
Count too: they bill as output.
Rule of thumb for English: is roughly 750 words.
With , the repeated start of each prompt is billed at the cheaper cache-read rate.
Gemini 3.1 Flash Lite
- Input: 72M tokens @ $0.25/M
- $18.00
- Cached input: 108M tokens @ $0.025/M
- $2.70
- Output: 37.5M tokens @ $1.5/M
- $56.25
- Total per month
- $76.95
Cost = tokens ÷ 1,000,000 × price per 1M, summed for input, cached input and output
Cheapest budget-tier model for this workload: DeepSeek V4.1 Flash at $23.93/month (69% less). Cheaper models can be weaker, so test them on your own prompts before switching.
This Workload on Every Model
180M input and 37.5M output tokens a month at real-time prices. Cheapest first; tap a model to load it above.
| Model | Tier | Input / 1M | Cached / 1M | Output / 1M | Context | Monthly cost | vs Gemini 3.1 Flash Lite |
|---|---|---|---|---|---|---|---|
| DeepSeek | Budget | $0.045 | $0.018 | $0.5 | 1M | $23.93 | -69% |
| Anthropic | Budget | $0.1 | $0.01 | $0.5 | 1M | $27.03 | -65% |
| OpenAI | Budget | $0.1 | $0.01 | $0.5 | 1.1M | $27.03 | -65% |
| Alibaba Qwen | Budget | $0.15 | $0.016 | $0.47 | 1M | $30.15 | -61% |
| Budget | $0.25 | $0.025 | $1.5 | 1M | $76.95 | selected | |
| Meta (open weights) | Budget | $0.35 | $0.17 | $1 | 524.3K | $81.06 | +5% |
| Mistral | Mid | $0.7 | $0.07 | $2.1 | 1M | $136.71 | +78% |
| DeepSeek | Mid | $0.103 | $0.098 | $4.11 | 1M | $172.13 | +124% |
| Mid | $0.825 | $0.0825 | $4.125 | 1M | $223.00 | +190% | |
| OpenAI | Mid | $1 | $0.05 | $5 | 1.1M | $264.90 | +244% |
| xAI | Flagship | $2 | $0.5 | $6 | 500K | $423.00 | +450% |
| Anthropic | Mid | $2 | $0.1 | $10 | 1M | $529.80 | +588% |
| Flagship | $2 | $0.1 | $10 | 1M | $529.80 | +588% | |
| OpenAI | Mid | $2 | $0.1 | $10 | 1.1M | $529.80 | +588% |
| Anthropic | Flagship | $4 | $0.2 | $20 | 1M | $1,059.60 | +1277% |
| Anthropic | Flagship | $10 | $0.25 | $50 | 1M | $2,622.00 | +3307% |
| OpenAI | Flagship | $10 | $1 | $50 | 1.1M | $2,703.00 | +3413% |
Prices are the per-token rates OpenRouter listed on 5 Nov 2026. Buying direct from a provider can be priced differently, and some providers charge more for very long prompts, so check the provider's pricing page before you commit a budget. Cache-write charges (paid the first time a prefix is cached) are not included; with a high cache share they add little. A month is 30 days.
Notes for This Preset
A support bot handling 5,000 requests a day (150,000 a month), with example inputs of 1,200 input tokens (FAQ knowledge base plus the customer's question) and 250 output tokens, costs $76.95/month ($0.513 per 1,000 requests) on Gemini 3.1 Flash Lite with 60% prompt caching, at the OpenRouter prices of 5 November 2026. On DeepSeek V4.1 Flash the same workload costs $23.93/month; on Claude Sonnet 5.5 it costs $529.80/month, so Flash Lite is about 85% cheaper. Cheaper models can answer worse, so test them on real tickets before you switch.
How can I lower the API bill for this workload?
Three levers, in order of effort: cache the repeated start of the prompt (system prompt, documents, tool definitions), move work that can wait to the Batch API, and test a cheaper model on your own prompts. Shorter answers help too, because output tokens cost several times more than input tokens.
Building an AI Chatbot, Agent or RAG App?
Vsurge Media builds production AI assistants, RAG search and automation for businesses, and helps keep the monthly API bill under control.