Developer Tools
AI API Cost Calculator
Estimate and compare API costs across OpenAI, Anthropic, Google, DeepSeek, Mistral, Qwen, and more. No sign-up required.
Start CalculatingConfigure Your Workload
Cost Breakdown — Gpt 4o
Input Cost$25.00
Output Cost$50.00
Cached Savings-$0.000
Total Monthly Cost$75.00
Avg Cost / Request$0.0075
Estimated Annual$900.00
Smart Insights
1
You could save approximately 98% by switching to Llama 3.1 8b (Meta), reducing monthly costs from $75.00 to $1.50.
2
This model supports cached input pricing. Adding cached tokens could reduce your input costs by up to 50%.
Cross-Model Comparison
Same workload (1,000 in / 500 out × 10,000 req/mo) across all supported models.
| Model | Provider | Input Cost | Output Cost | Monthly Total | vs Selected |
|---|---|---|---|---|---|
Llama 3.1 8bCheapest | Meta | $1.00 | $0.500 | $1.50 | -98% |
Gemini 1.5 Flash | $0.750 | $1.50 | $2.25 | -97% | |
Deepseek V3 | DeepSeek | $1.40 | $1.40 | $2.80 | -96% |
Gemini 2.0 Flash | $1.00 | $2.00 | $3.00 | -96% | |
Gpt 4o Mini | OpenAI | $1.50 | $3.00 | $4.50 | -94% |
Mistral Small | Mistral | $2.00 | $3.00 | $5.00 | -93% |
Codestral | Mistral | $2.00 | $3.00 | $5.00 | -93% |
Qwen 2.5 72b | Qwen | $3.50 | $2.00 | $5.50 | -93% |
Claude 3 Haiku | Anthropic | $2.50 | $6.25 | $8.75 | -88% |
Llama 3.3 70b | Meta | $6.00 | $3.00 | $9.00 | -88% |
Llama 3.1 70b | Meta | $6.00 | $3.00 | $9.00 | -88% |
Gpt 3.5 Turbo | OpenAI | $5.00 | $7.50 | $12.50 | -83% |
Deepseek R1 | DeepSeek | $5.50 | $10.95 | $16.45 | -78% |
Claude 3 5 Haiku | Anthropic | $8.00 | $20.00 | $28.00 | -63% |
Gemini 1.5 Pro | $12.50 | $25.00 | $37.50 | -50% | |
Llama 3.1 405b | Meta | $30.00 | $15.00 | $45.00 | -40% |
Qwen 2.5 Max | Qwen | $16.00 | $32.00 | $48.00 | -36% |
Mistral Large 2 | Mistral | $20.00 | $30.00 | $50.00 | -33% |
Gpt 4oSelected | OpenAI | $25.00 | $50.00 | $75.00 | — |
O1 Mini | OpenAI | $30.00 | $60.00 | $90.00 | +20% |
Claude 3 5 Sonnet | Anthropic | $30.00 | $75.00 | $105.00 | +40% |
Gpt 4 Turbo | OpenAI | $100.00 | $150.00 | $250.00 | +233% |
O1 | OpenAI | $150.00 | $300.00 | $450.00 | +500% |
Claude 3 Opus | Anthropic | $150.00 | $375.00 | $525.00 | +600% |
Llama 3.1 8bCheapest
MetaInput$1.00
Output$0.500
Monthly$1.50
Gemini 1.5 Flash
GoogleInput$0.750
Output$1.50
Monthly$2.25
Deepseek V3
DeepSeekInput$1.40
Output$1.40
Monthly$2.80
Gemini 2.0 Flash
GoogleInput$1.00
Output$2.00
Monthly$3.00
Gpt 4o Mini
OpenAIInput$1.50
Output$3.00
Monthly$4.50
Mistral Small
MistralInput$2.00
Output$3.00
Monthly$5.00
Codestral
MistralInput$2.00
Output$3.00
Monthly$5.00
Qwen 2.5 72b
QwenInput$3.50
Output$2.00
Monthly$5.50
Claude 3 Haiku
AnthropicInput$2.50
Output$6.25
Monthly$8.75
Llama 3.3 70b
MetaInput$6.00
Output$3.00
Monthly$9.00
Llama 3.1 70b
MetaInput$6.00
Output$3.00
Monthly$9.00
Gpt 3.5 Turbo
OpenAIInput$5.00
Output$7.50
Monthly$12.50
Deepseek R1
DeepSeekInput$5.50
Output$10.95
Monthly$16.45
Claude 3 5 Haiku
AnthropicInput$8.00
Output$20.00
Monthly$28.00
Gemini 1.5 Pro
GoogleInput$12.50
Output$25.00
Monthly$37.50
Llama 3.1 405b
MetaInput$30.00
Output$15.00
Monthly$45.00
Qwen 2.5 Max
QwenInput$16.00
Output$32.00
Monthly$48.00
Mistral Large 2
MistralInput$20.00
Output$30.00
Monthly$50.00
Gpt 4oSelected
OpenAIInput$25.00
Output$50.00
Monthly$75.00
O1 Mini
OpenAIInput$30.00
Output$60.00
Monthly$90.00
Claude 3 5 Sonnet
AnthropicInput$30.00
Output$75.00
Monthly$105.00
Gpt 4 Turbo
OpenAIInput$100.00
Output$150.00
Monthly$250.00
O1
OpenAIInput$150.00
Output$300.00
Monthly$450.00
Claude 3 Opus
AnthropicInput$150.00
Output$375.00
Monthly$525.00
Relative Cost Comparison
Llama 3.1 8b
$1.50
Gemini 1.5 Flash
$2.25
Deepseek V3
$2.80
Gemini 2.0 Flash
$3.00
Gpt 4o Mini
$4.50
Mistral Small
$5.00
Codestral
$5.00
Qwen 2.5 72b
$5.50
Claude 3 Haiku
$8.75
Llama 3.3 70b
$9.00
Llama 3.1 70b
$9.00
Gpt 3.5 Turbo
$12.50
Frequently Asked Questions
Common questions about AI API pricing and how this calculator works.
Explore More
Related Developer Tools
Coming Soon
Token Calculator
Compute BPE token counts from raw text.
Coming Soon
Context Calculator
Calculate context fill ratios for long-context models.
Coming Soon
VRAM Calculator
GPU requirements for local model inference.