Token Counter & Cost Calculator forLLM APIs
Compare token usage and pricing across OpenAI, Anthropic, Google, Mistral, AWS Bedrock, and many more. Estimate project costs before you write a single line of code.
100% Private: All calculations happen in your browser. Your text is never sent to any server or stored anywhere.
Paste your full prompt including system messages and context to get accurate token counts.
Enter the expected average token count for AI responses. Typical responses are 100-2000 tokens.
Average length of AI response
Mistral | Cost optimized for your usage
Model Comparison Matrix
Ministral 3B | Mistral | 128K | $0.040 | $0.040 | $0.00002000 | $0.6000 |
Llama 3.1 8B (Groq) | Groq | 128K | $0.050 | $0.080 | $0.00004000 | $1.20 |
Ministral 8B | Mistral | 128K | $0.100 | $0.100 | $0.00005000 | $1.50 |
Nova Micro | AWS Bedrock | 128K | $0.035 | $0.140 | $0.00007000 | $2.10 |
Titan Text Lite | AWS Bedrock | 4K | $0.150 | $0.200 | $0.000100 | $3.00 |
Qwen 2.5 32B | Qwen | 128K | $0.150 | $0.200 | $0.000100 | $3.00 |
Qwen 2.5 Coder 32B | Qwen | 128K | $0.150 | $0.200 | $0.000100 | $3.00 |
Nova LitePopular | AWS Bedrock | 300K | $0.060 | $0.240 | $0.000120 | $3.60 |
Mixtral 8x7B (Groq) | Groq | 32K | $0.240 | $0.240 | $0.000120 | $3.60 |
DeepSeek Coder V2 | DeepSeek | 128K | $0.140 | $0.280 | $0.000140 | $4.20 |
Gemini 3.1 FlashPopular | 1.0M | $0.075 | $0.300 | $0.000150 | $4.50 | |
Gemini 3 FlashPopular | 1.0M | $0.075 | $0.300 | $0.000150 | $4.50 | |
Gemini 2.5 Flash | 1.0M | $0.075 | $0.300 | $0.000150 | $4.50 | |
Llama 4 ScoutPopular | Meta | 1.0M | $0.100 | $0.300 | $0.000150 | $4.50 |
GPT-5 Nano | OpenAI | 128K | $0.050 | $0.400 | $0.000200 | $6.00 |
GPT-4.1 Nano | OpenAI | 1.0M | $0.100 | $0.400 | $0.000200 | $6.00 |
Gemini 2.0 FlashPopular | 1.0M | $0.100 | $0.400 | $0.000200 | $6.00 | |
Qwen 2.5 72BFlagship | Qwen | 128K | $0.350 | $0.400 | $0.000200 | $6.00 |
Jamba 1.5 Mini | AI21 | 256K | $0.200 | $0.400 | $0.000200 | $6.00 |
GPT-4o MiniPopular | OpenAI | 128K | $0.150 | $0.600 | $0.000300 | $9.00 |
Mistral SmallPopular | Mistral | 32K | $0.200 | $0.600 | $0.000300 | $9.00 |
Titan Text ExpressPopular | AWS Bedrock | 8K | $0.200 | $0.600 | $0.000300 | $9.00 |
QwQ 32B | Qwen | 32K | $0.150 | $0.600 | $0.000300 | $9.00 |
Llama 4 MaverickFlagship | Meta | 1.0M | $0.200 | $0.600 | $0.000300 | $9.00 |
Llama 3.3 70B | Meta | 128K | $0.590 | $0.790 | $0.000395 | $11.85 |
Llama 3.3 70B (Groq)Popular | Groq | 128K | $0.590 | $0.790 | $0.000395 | $11.85 |
Titan Multimodal | AWS Bedrock | 8K | $0.800 | $0.800 | $0.000400 | $12.00 |
Codestral | Mistral | 32K | $0.300 | $0.900 | $0.000450 | $13.50 |
DeepSeek V3Popular | DeepSeek | 64K | $0.270 | $1.10 | $0.000550 | $16.50 |
Titan Text PremierFlagship | AWS Bedrock | 32K | $0.500 | $1.50 | $0.000750 | $22.50 |
Command RPopular | Cohere | 128K | $0.500 | $1.50 | $0.000750 | $22.50 |
GPT-4.1 MiniPopular | OpenAI | 1.0M | $0.400 | $1.60 | $0.000800 | $24.00 |
GPT-5 MiniPopular | OpenAI | 128K | $0.250 | $2.00 | $0.001000 | $30.00 |
DeepSeek R1Flagship | DeepSeek | 64K | $0.550 | $2.19 | $0.001095 | $32.85 |
Nova 2 Lite | AWS Bedrock | 1.0M | $0.300 | $2.50 | $0.001250 | $37.50 |
Nova ProFlagship | AWS Bedrock | 300K | $0.800 | $3.20 | $0.001600 | $48.00 |
Claude 3.5 HaikuPopular | Anthropic | 200K | $0.800 | $4.00 | $0.002000 | $60.00 |
Qwen VL Max | Qwen | 32K | $2.00 | $4.00 | $0.002000 | $60.00 |
o4-miniPopular | OpenAI | 200K | $1.10 | $4.40 | $0.002200 | $66.00 |
o3-mini | OpenAI | 200K | $1.10 | $4.40 | $0.002200 | $66.00 |
o1-mini | OpenAI | 128K | $1.10 | $4.40 | $0.002200 | $66.00 |
Gemini 3.1 ProFlagship | 2.0M | $1.25 | $5.00 | $0.002500 | $75.00 | |
Gemini 3 ProFlagship | 2.0M | $1.25 | $5.00 | $0.002500 | $75.00 | |
Gemini 2.5 Pro +1 | 2.0M | $1.25 | $5.00 | $0.002500 | $75.00 | |
Codex Mini | OpenAI | 200K | $1.50 | $6.00 | $0.003000 | $90.00 |
Mistral LargeFlagship | Mistral | 128K | $2.00 | $6.00 | $0.003000 | $90.00 |
Pixtral Large | Mistral | 128K | $2.00 | $6.00 | $0.003000 | $90.00 |
Grok 2 Mini | xAI | 128K | $2.00 | $6.00 | $0.003000 | $90.00 |
GPT-4.1 | OpenAI | 1.0M | $2.00 | $8.00 | $0.004000 | $120.00 |
Jamba 1.5 LargeFlagship | AI21 | 256K | $2.00 | $8.00 | $0.004000 | $120.00 |
GPT-5.1 | OpenAI | 200K | $1.25 | $10.00 | $0.005000 | $150.00 |
GPT-5Flagship | OpenAI | 200K | $1.25 | $10.00 | $0.005000 | $150.00 |
GPT-4o | OpenAI | 128K | $2.50 | $10.00 | $0.005000 | $150.00 |
Command R+Flagship | Cohere | 128K | $2.50 | $10.00 | $0.005000 | $150.00 |
GPT-5.2Flagship | OpenAI | 200K | $1.75 | $14.00 | $0.007000 | $210.00 |
GPT-5.4Flagship | OpenAI | 272K | $2.50 | $15.00 | $0.007500 | $225.00 |
Claude 4.6 SonnetPopular +1 | Anthropic | 200K | $3.00 | $15.00 | $0.007500 | $225.00 |
Claude 4 Sonnet +1 | Anthropic | 200K | $3.00 | $15.00 | $0.007500 | $225.00 |
Claude 3.7 Sonnet +2 | Anthropic | 200K | $3.00 | $15.00 | $0.007500 | $225.00 |
Claude 3.5 SonnetPopular +1 | Anthropic | 200K | $3.00 | $15.00 | $0.007500 | $225.00 |
Grok 2Flagship | xAI | 128K | $5.00 | $15.00 | $0.007500 | $225.00 |
Nova 2 ProFlagship | AWS Bedrock | 1.0M | $2.19 | $17.50 | $0.008750 | $262.50 |
o3Flagship | OpenAI | 200K | $10.00 | $40.00 | $0.0200 | $600.00 |
o1 | OpenAI | 200K | $15.00 | $60.00 | $0.0300 | $900.00 |
Claude 4.6 OpusFlagship | Anthropic | 200K | $15.00 | $75.00 | $0.0375 | $1.13K |
Claude 4 OpusFlagship | Anthropic | 200K | $15.00 | $75.00 | $0.0375 | $1.13K |
Claude 3 Opus | Anthropic | 200K | $15.00 | $75.00 | $0.0375 | $1.13K |
o3 ProFlagship | OpenAI | 200K | $20.00 | $80.00 | $0.0400 | $1.20K |
GPT-5.4 ProFlagship | OpenAI | 272K | $30.00 | $180.00 | $0.0900 | $2.70K |
How Token Pricing Works
Understanding token-based billing helps you make smarter decisions about which models to use for your applications.
Token Counting
LLM providers charge based on the number of tokens processed. Tokens are pieces of text - roughly 4 characters or 0.75 words on average.
Input vs Output Tokens
Input tokens are what you send to the model (prompts, context). Output tokens are the generated response. Each has different pricing.
Tiered Pricing
Different models have vastly different costs. Flagship models like GPT-4o cost more than efficient alternatives like GPT-4o-mini.
Cost Optimization
By comparing costs across providers, you can save 10-90% on your LLM spending without sacrificing quality for your use case.
The Calculation Formula
Rates are typically expressed per 1 million tokens (1M). Our calculator automatically applies the correct rates for each provider and multiplies by your expected usage volume.
Pro Tip
You can paste both your input prompt AND an example output response to get accurate token counts for both. This gives you the most precise cost estimate for your specific use case.