LLM API Pricing Calculator 2026: Live Prices for 300+ Models
LLM API Pricing Calculator
Compare what every major LLM API costs per million tokens — GPT, Claude, Gemini, Llama, DeepSeek and 300+ more — and estimate your monthly bill. Live prices refresh every 6 hours from OpenRouter's public catalog.
Live LLM API price table
● Live data — updates automaticallyPrices refresh every 6 hours from OpenRouter's public API.
| Model ⇅ | $ / 1M input ⇅ | $ / 1M output ⇅ | Context ⇅ |
|---|---|---|---|
| Loading live prices… (if this doesn't load, see the reference table below) | |||
Click a column header to sort. Prices in USD per 1 million tokens.
Monthly cost calculator
Estimated monthly cost = input millions × input price + output millions × output price. Excludes prompt-caching discounts and batch pricing.
Token counter
Rough estimate: ~1 token per 4 characters of English text. Exact counts vary by model tokenizer.
Popular LLM API prices at a glance
Hand-curated reference — last verified July 16, 2026. Indicative list prices in USD per 1M tokens; the live table above is the up-to-date source. Always confirm on the provider's pricing page before committing to volume.
| Model | $ / 1M input | $ / 1M output | Context |
|---|---|---|---|
| Anthropic Claude Fable 5 | $10.00 | $50.00 | 1M |
| Anthropic Claude Opus 4.8 | $5.00 | $25.00 | 1M |
| Anthropic Claude Sonnet 5 | $2.00 | $10.00 | 1M |
| Anthropic Claude Haiku 4.5 | $1.00 | $5.00 | 200K |
| OpenAI GPT-5 | $1.25 | $10.00 | 400K |
| OpenAI GPT-5 Mini | $0.25 | $2.00 | 400K |
| OpenAI GPT-5 Nano | $0.05 | $0.40 | 400K |
| OpenAI GPT-4o | $2.50 | $10.00 | 128K |
| OpenAI GPT-4.1 | $2.00 | $8.00 | 1M |
| Google Gemini 3.1 Pro | $2.00 | $12.00 | 1M |
| Google Gemini 2.5 Pro | $1.25 | $10.00 | 1M |
| Google Gemini 2.5 Flash | $0.30 | $2.50 | 1M |
| Google Gemini 3.1 Flash Lite | $0.25 | $1.50 | 1M |
| xAI Grok 4.5 | $2.00 | $6.00 | 500K |
| xAI Grok 4.20 | $1.25 | $2.50 | 2M |
| DeepSeek V3.2 | $0.27 | $0.40 | 164K |
| Meta Llama 4 Maverick | $0.20 | $0.80 | 1M |
| Meta Llama 3.3 70B | $0.13 | $0.40 | 131K |
| Qwen3 Coder 480B | $0.30 | $1.00 | 1M |
| Moonshot Kimi K2 | $0.60 | $2.50 | 262K |
LLM API pricing FAQ
What does the GPT API cost?
OpenAI's GPT-5 costs $1.25 per 1M input tokens and $10 per 1M output tokens. The smaller GPT-5 Mini is $0.25 / $2.00, and GPT-5 Nano is $0.05 / $0.40 per 1M tokens. The older GPT-4o remains at $2.50 / $10.00. A typical chat request of 1,000 input and 500 output tokens on GPT-5 costs about $0.006.
How much does the Claude API cost?
Anthropic prices Claude by tier: Claude Haiku 4.5 is $1 / $5 per 1M tokens (input/output), Claude Sonnet 5 is $2 / $10 (introductory pricing), Claude Opus 4.8 is $5 / $25, and the flagship Claude Fable 5 is $10 / $50. Prompt caching can cut input costs by up to 90% for repeated context, and batch processing halves prices for non-urgent workloads.
What is the cheapest LLM API?
Among capable models, open-weight ones served via API are cheapest: Llama 3.3 70B (~$0.13 / $0.40 per 1M tokens), Llama 4 Maverick (~$0.20 / $0.80), and DeepSeek V3.2 (~$0.27 / $0.40). From the big labs, GPT-5 Nano ($0.05 / $0.40) and Gemini 2.5 Flash Lite ($0.10 / $0.40) are the budget picks. Some providers also offer free rate-limited tiers of open models — use the live table above and sort by input price to find them.
What is a token?
A token is the unit LLMs read and write text in — roughly 4 characters or ¾ of an English word. "LLM pricing is per token" is about 8 tokens. A page of text is ~500 tokens, and a 100-page document is ~50,000 tokens. Every API bills you per token processed, so both your prompt (input) and the model's reply (output) count toward cost.
What's the difference between input and output tokens?
Input tokens are everything you send to the model: system prompt, conversation history, documents, the user's question. Output tokens are what the model generates in response. Output tokens typically cost 3–5× more than input tokens because generation is more compute-intensive than reading. Most real workloads are input-heavy (long context, short answers), so input price usually dominates your bill.
Related tools
Planning to self-host instead? Check which GPU you need to run an LLM and compare cloud GPU prices. Building agents? Browse the MCP server directory.
