Monthly cost calculator
Choose a preset scenario or enter your own numbers. Costs use live list prices (no caching discounts) for the featured models.
| # | |||
|---|---|---|---|
| 1 | Qwen: Qwen3.7 Flash | $0.760 | — |
| 2 | DeepSeek: DeepSeek V4.1 Flash | $1.89 | 2.5× |
| 3 | OpenAI: GPT-5 Nano | $2.00 | 2.6× |
| 4 | Google: Gemini 2.5 Flash Lite | $2.40 | 3.2× |
| 5 | Anthropic: Claude Haiku 5.5 | $2.80 | 3.7× |
| 6 | OpenAI: GPT-6 Luna Pro | $2.80 | 3.7× |
| 7 | OpenAI: GPT-6 Luna | $2.80 | 3.7× |
| 8 | Qwen: Qwen3.8 Omni Flash | $3.08 | 4.1× |
| 9 | Qwen: Qwen3.8 Flash | $3.08 | 4.1× |
| 10 | DeepSeek: DeepSeek V4.1 Flash 1017 | $3.60 | 4.7× |
| 11 | Meta: Llama 4 Maverick | $4.11 | 5.4× |
| 12 | DeepSeek: DeepSeek V4 Flash Vision Exp | $4.31 | 5.7× |
| 13 | Qwen: Qwen3.8 27B | $5.72 | 7.5× |
| 14 | DeepSeek: DeepSeek V3.1 | $5.80 | 7.6× |
| 15 | OpenAI: GPT-5.6 Luna | $6.40 | 8.4× |
Cheapest for this workload: Qwen: Qwen3.7 Flash at $0.760/mo.
🔥 Have a real bill? Shred it →Frequently asked questions
- How is monthly LLM cost calculated?
- Monthly cost = requests × (input tokens × input price + output tokens × output price) ÷ 1,000,000. The calculator applies this formula to live list prices for every featured model.
- Do these estimates include prompt caching discounts?
- No — figures use standard list prices without caching. Prompt caching can cut input costs by up to 90% on repeated context, so real bills are often lower than these estimates.
- Which model is cheapest for my workload?
- It depends on your input/output token ratio. Input-heavy workloads (RAG, coding agents) favor cheap-input models; balanced workloads should compare blended cost. The calculator ranks all models for your exact numbers.
- How current are the prices?
- Prices are fetched from OpenRouter's public API and rebuilt daily. Each model page shows its verification date and links to the provider source.