AI API token prices
and request calculator
Compare input and output token prices, calculate a real workload and track provider changes.
Choose a typical workload or enter your own values. Prices in the table are recalculated instantly.
Use the + buttons next to model names.
| Model | Input / 1M | Output / 1M | Change | Your request |
|---|---|---|---|---|
| OpenAI: o1-pro⌄OpenaiLegacyVery expensive | $150 | $600 | 0% | — |
MODEL PROFILE OpenAI: o1-proAPI profile based on provider metadata: image input, file input, coding, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingComplex analysisStructured outputLong contextLong responses
| ||||
| OpenAI: GPT-5.5 Pro⌄OpenaiVery expensive | $30 | $180 | 0% | — |
MODEL PROFILE OpenAI: GPT-5.5 ProAPI profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| OpenAI: GPT-5.4 Pro⌄OpenaiVery expensive | $30 | $180 | 0% | — |
MODEL PROFILE OpenAI: GPT-5.4 ProAPI profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| OpenAI: GPT-5.2 Pro⌄OpenaiVery expensive | $21 | $168 | 0% | — |
MODEL PROFILE OpenAI: GPT-5.2 ProAPI profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| OpenAI: GPT-5 Pro⌄OpenaiVery expensive | $15 | $120 | 0% | — |
MODEL PROFILE OpenAI: GPT-5 ProAPI profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| OpenAI: o3 Pro⌄OpenaiVery expensive | $20 | $80 | 0% | — |
MODEL PROFILE OpenAI: o3 ProAPI profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Anthropic: Claude Opus 4.1⌄AnthropicVery expensive | $15 | $75 | 0% | — |
MODEL PROFILE Anthropic: Claude Opus 4.1API profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisLong contextLong responses
| ||||
| OpenAI: GPT-4⌄OpenaiVery expensive | $30 | $60 | 0% | — |
MODEL PROFILE OpenAI: GPT-4API profile based on provider metadata: ai agents & tools, structured output. AI agents & toolsStructured output
| ||||
| OpenAI: o1⌄OpenaiVery expensive | $15 | $60 | 0% | — |
MODEL PROFILE OpenAI: o1API profile based on provider metadata: image input, file input, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| OpenAI: GPT-6 Astra⌄OpenaiFlagshipVery expensive | $10 | $50 | 0% | — |
MODEL PROFILE OpenAI: GPT-6 AstraAPI profile based on provider metadata: image input, file input, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| OpenAI: GPT-6 Astra Pro⌄OpenaiFlagshipVery expensive | $10 | $50 | 0% | — |
MODEL PROFILE OpenAI: GPT-6 Astra ProAPI profile based on provider metadata: image input, file input, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Anthropic: Claude Fable 5.1⌄AnthropicFlagshipVery expensive | $10 | $50 | 0% | — |
MODEL PROFILE Anthropic: Claude Fable 5.1API profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Anthropic: Claude Fable 5⌄AnthropicFlagshipVery expensive | $10 | $50 | 0% | — |
MODEL PROFILE Anthropic: Claude Fable 5API profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| OpenAI: GPT-4 Turbo⌄Openai | $10 | $30 | 0% | — |
MODEL PROFILE OpenAI: GPT-4 TurboAPI profile based on provider metadata: image input, ai agents & tools, structured output. Image inputAI agents & toolsStructured output
| ||||
| OpenAI: GPT Chat Latest⌄Openai | $5 | $30 | 0% | — |
MODEL PROFILE OpenAI: GPT Chat LatestAPI profile based on provider metadata: image input, file input, ai agents & tools, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputAI agents & toolsStructured outputLong contextLong responses
| ||||
| OpenAI: GPT-5.5⌄Openai | $5 | $30 | 0% | — |
MODEL PROFILE OpenAI: GPT-5.5API profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Anthropic: Claude Opus 5⌄AnthropicFlagship | $5 | $25 | 0% | — |
MODEL PROFILE Anthropic: Claude Opus 5API profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Anthropic: Claude Opus 4.8⌄Anthropic | $5 | $25 | 0% | — |
MODEL PROFILE Anthropic: Claude Opus 4.8API profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Anthropic: Claude Opus 4.7⌄Anthropic | $5 | $25 | 0% | — |
MODEL PROFILE Anthropic: Claude Opus 4.7API profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Anthropic: Claude Opus 4.6⌄Anthropic | $5 | $25 | 0% | — |
MODEL PROFILE Anthropic: Claude Opus 4.6API profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Anthropic: Claude Opus 4.5⌄Anthropic | $5 | $25 | 0% | — |
MODEL PROFILE Anthropic: Claude Opus 4.5API profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Anthropic: Claude Opus 5.5⌄AnthropicFlagship | $4 | $20 | 0% | — |
MODEL PROFILE Anthropic: Claude Opus 5.5API profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| OpenAI: GPT-5.6 Sol Pro⌄Openai | $4 | $20 | 0% | — |
MODEL PROFILE OpenAI: GPT-5.6 Sol ProAPI profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| OpenAI: GPT-5.4 Image 2⌄Openai | $8 | $15 | 0% | — |
MODEL PROFILE OpenAI: GPT-5.4 Image 2API profile based on provider metadata: image input, file input, image generation, coding, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputImage generationCodingComplex analysisStructured outputLong contextLong responses
| ||||
| OpenAI: GPT-5 Image⌄Openai | $10 | $10 | 0% | — |
MODEL PROFILE OpenAI: GPT-5 ImageAPI profile based on provider metadata: image input, file input, image generation, coding, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputImage generationCodingComplex analysisStructured outputLong contextLong responses
| ||||
| OpenAI: GPT-4o (2024-05-13)⌄Openai | $5 | $15 | 0% | — |
MODEL PROFILE OpenAI: GPT-4o (2024-05-13)API profile based on provider metadata: image input, file input, ai agents & tools, structured output. Image inputFile inputAI agents & toolsStructured output
| ||||
| Qwen: Qwen3.8 Max Prime⌄Qwen | $4 | $12 | 0% | — |
MODEL PROFILE Qwen: Qwen3.8 Max PrimeAPI profile based on provider metadata: image input, video input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputVideo inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Anthropic: Claude Sonnet 5.5⌄Anthropic | $2 | $10 | 0% | — |
MODEL PROFILE Anthropic: Claude Sonnet 5.5API profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| SpaceXAI: Grok 4.7⌄X-ai | $2 | $6 | 0% | — |
MODEL PROFILE SpaceXAI: Grok 4.7API profile based on provider metadata: image input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Qwen: Qwen3.8 Max (0902)⌄Qwen | $2 | $6 | 0% | — |
MODEL PROFILE Qwen: Qwen3.8 Max (0902)API profile based on provider metadata: image input, video input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputVideo inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Qwen: Qwen3.8 2.4T A95B⌄Qwen | $2 | $6 | 0% | — |
MODEL PROFILE Qwen: Qwen3.8 2.4T A95BAPI profile based on provider metadata: coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. CodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Google: Gemini 3.8 Flash⌄Google | $0.75 | $3.75 | 0% | — |
MODEL PROFILE Google: Gemini 3.8 FlashAPI profile based on provider metadata: image input, audio, video input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputAudioVideo inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Google: Gemini 3.7 Flash⌄Google | $0.75 | $3.75 | 0% | — |
MODEL PROFILE Google: Gemini 3.7 FlashAPI profile based on provider metadata: image input, audio, video input, file input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputAudioVideo inputFile inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Qwen: Qwen3.8 27B⌄Qwen | $0.42 | $3 | 0% | — |
MODEL PROFILE Qwen: Qwen3.8 27BAPI profile based on provider metadata: image input, video input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputVideo inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| DeepSeek: DeepSeek V4 Pro 0813⌄Deepseek | $0.66 | $1.98 | 0% | — |
MODEL PROFILE DeepSeek: DeepSeek V4 Pro 0813API profile based on provider metadata: coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. CodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| DeepSeek: DeepSeek V4 Flash Vision Exp⌄DeepseekGood value | $0.2156 | $0.6468 | 0% | — |
MODEL PROFILE DeepSeek: DeepSeek V4 Flash Vision ExpAPI profile based on provider metadata: image input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Qwen: Qwen3.8 Omni Flash⌄QwenGood value | $0.15 | $0.47 | 0% | — |
MODEL PROFILE Qwen: Qwen3.8 Omni FlashAPI profile based on provider metadata: image input, audio, video input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputAudioVideo inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Qwen: Qwen3.8 Flash⌄QwenGood value | $0.15 | $0.47 | 0% | — |
MODEL PROFILE Qwen: Qwen3.8 FlashAPI profile based on provider metadata: image input, video input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputVideo inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| DeepSeek: DeepSeek V4.1 Flash⌄DeepseekGood value | $0.03 | $0.5 | 0% | — |
MODEL PROFILE DeepSeek: DeepSeek V4.1 FlashAPI profile based on provider metadata: image input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Qwen: Qwen3.5-Flash⌄QwenGood value | $0.065 | $0.26 | 0% | — |
MODEL PROFILE Qwen: Qwen3.5-FlashAPI profile based on provider metadata: image input, video input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputVideo inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Qwen: Qwen2.5 7B Instruct⌄QwenGood value | $0.1 | $0.2 | 0% | — |
MODEL PROFILE Qwen: Qwen2.5 7B InstructAPI profile based on provider metadata: coding, ai agents & tools, structured output. CodingAI agents & toolsStructured output
| ||||
| Mistral: Ministral 3 8B 2512⌄MistralaiGood value | $0.15 | $0.15 | 0% | — |
MODEL PROFILE Mistral: Ministral 3 8B 2512API profile based on provider metadata: image input, ai agents & tools, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputAI agents & toolsStructured outputLong contextLong responses
| ||||
| Qwen: Qwen3.5-9B⌄QwenGood value | $0.1 | $0.15 | 0% | — |
MODEL PROFILE Qwen: Qwen3.5-9BAPI profile based on provider metadata: image input, video input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputVideo inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Qwen: Qwen3 30B A3B Instruct 2507⌄QwenGood value | $0.0482 | $0.1931 | 0% | — |
MODEL PROFILE Qwen: Qwen3 30B A3B Instruct 2507API profile based on provider metadata: coding, ai agents & tools, structured output, long context, long responses. Suitable for workloads with a large context window. CodingAI agents & toolsStructured outputLong contextLong responses
| ||||
| Meta: Llama 3.2 1B Instruct⌄Meta-llamaGood value | $0.027 | $0.201 | 0% | — |
MODEL PROFILE Meta: Llama 3.2 1B InstructAPI profile based on provider metadata: long responses. Long responses
| ||||
| OpenAI: gpt-oss-120b⌄OpenaiGood value | $0.037 | $0.17 | 0% | — |
MODEL PROFILE OpenAI: gpt-oss-120bAPI profile based on provider metadata: ai agents & tools, complex analysis, structured output, long responses. AI agents & toolsComplex analysisStructured outputLong responses
| ||||
| Google: Gemma 3 12B⌄GoogleGood value | $0.05 | $0.15 | 0% | — |
MODEL PROFILE Google: Gemma 3 12BAPI profile based on provider metadata: image input, ai agents & tools, structured output. Image inputAI agents & toolsStructured output
| ||||
| Mistral: Ministral 3 3B 2512⌄MistralaiGood value | $0.1 | $0.1 | 0% | — |
MODEL PROFILE Mistral: Ministral 3 3B 2512API profile based on provider metadata: image input, ai agents & tools, structured output, long responses. Image inputAI agents & toolsStructured outputLong responses
| ||||
| Qwen: Qwen3.7 Flash⌄QwenGood value | $0.03 | $0.13 | 0% | — |
MODEL PROFILE Qwen: Qwen3.7 FlashAPI profile based on provider metadata: image input, video input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputVideo inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Google: Gemma 3 4B⌄GoogleGood value | $0.05 | $0.1 | 0% | — |
MODEL PROFILE Google: Gemma 3 4BAPI profile based on provider metadata: image input, structured output. Image inputStructured output
| ||||
| Meta: Llama 3.1 8B Instruct⌄Meta-llamaGood value | $0.05 | $0.08 | 0% | — |
MODEL PROFILE Meta: Llama 3.1 8B InstructAPI profile based on provider metadata: ai agents & tools, structured output, long responses. AI agents & toolsStructured outputLong responses
| ||||
| Mistral: Mistral Small 3⌄MistralaiGood value | $0.05 | $0.08 | 0% | — |
MODEL PROFILE Mistral: Mistral Small 3API profile based on provider metadata: structured output. Structured output
| ||||
| DeepSeek: DeepSeek V4 Flash 0423⌄DeepseekGood value | $0.0419 | $0.0837 | 0% | — |
MODEL PROFILE DeepSeek: DeepSeek V4 Flash 0423API profile based on provider metadata: coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. CodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| OpenAI: gpt-oss-20b⌄OpenaiGood value | $0.018 | $0.09 | 0% | — |
MODEL PROFILE OpenAI: gpt-oss-20bAPI profile based on provider metadata: ai agents & tools, complex analysis, structured output, long responses. AI agents & toolsComplex analysisStructured outputLong responses
| ||||
| Mistral: Mistral Nemo⌄MistralaiGood value | $0.019 | $0.03 | 0% | — |
MODEL PROFILE Mistral: Mistral NemoAPI profile based on provider metadata: ai agents & tools, structured output. AI agents & toolsStructured output
| ||||
| Google: Lyria 3 Clip Preview⌄GoogleFree | $0 | $0 | — | — |
MODEL PROFILE Google: Lyria 3 Clip PreviewAPI profile based on provider metadata: image input, audio, coding, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputAudioCodingStructured outputLong contextLong responses
| ||||
| Google: Lyria 3 Pro Preview⌄GoogleFree | $0 | $0 | — | — |
MODEL PROFILE Google: Lyria 3 Pro PreviewAPI profile based on provider metadata: image input, audio, coding, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputAudioCodingStructured outputLong contextLong responses
| ||||
| Google: Gemma 4 31B (free)⌄GoogleFree | $0 | $0 | — | — |
MODEL PROFILE Google: Gemma 4 31B (free)API profile based on provider metadata: image input, video input, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputVideo inputAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Google: Gemma 4 26B A4B (free)⌄GoogleFree | $0 | $0 | — | — |
MODEL PROFILE Google: Gemma 4 26B A4B (free)API profile based on provider metadata: image input, video input, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputVideo inputAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
| Qwen: Qwen3.8 27B (free)⌄QwenFree | $0 | $0 | — | — |
MODEL PROFILE Qwen: Qwen3.8 27B (free)API profile based on provider metadata: image input, video input, coding, ai agents & tools, complex analysis, structured output, long context, long responses. Suitable for workloads with a large context window. Image inputVideo inputCodingAI agents & toolsComplex analysisStructured outputLong contextLong responses
| ||||
Estimates use currently published token rates. Caching, reasoning, tools, images and provider fees may change the final bill. Source: OpenRouter model API; official provider pages are used for editorial verification.
Compare providers and real workloads
How to compare AI API token prices
AI API pricing usually has separate rates for input tokens sent to a model and output tokens generated in the response. Enter your typical prompt, answer length and number of requests in the calculator to compare GPT, Claude, Gemini, DeepSeek, Llama, Mistral and other models on the same workload.
Price per 1M tokens
The standard unit used by providers. The table keeps input and output rates separate so expensive generations do not disappear behind a low prompt price.
Estimated request cost
The calculator multiplies your input and output volume by the current rates. It is useful for prototypes, chatbots, agents, content pipelines and production forecasts.
Hourly price monitoring
The catalogue is checked every hour. The update time shown above tells you when the current snapshot was collected. Historical points accumulate from actual checks.
Frequently asked questions
What is an AI token?
A token is a text unit processed by a model. One token may be a word, part of a word, punctuation mark or other fragment, depending on the tokenizer.
Why are input and output prices different?
Generating new tokens generally requires more computation than reading a prompt, so providers often charge more for output.
Does the calculator show the final provider bill?
It provides an estimate based on published token rates. Caching, reasoning tokens, images, tools, minimum charges, routing and taxes may affect the final amount.
How often are AI model prices updated?
AI Feed checks the catalogue hourly. The timestamp at the top of the page shows the latest successful data snapshot.