LLM Cost

Price rankings

Cheapest LLM APIs right now

Paid models ranked from the live catalog by list price per million tokens. Compare the input and output rankings separately because the cheapest prompt model is not always the cheapest model for generated text.

1,265 published models - 79 providers - refreshed hourly

Lowest input cost per 1M tokens

Useful when prompts, retrieval context, or documents dominate the workload.

  1. #1Flux 1 Dev Controlnet UnionFireworks Ai - 4K context$0.0010/M
  2. #2Ling-2.6-flashInclusionai - 262K context$0.010/M
  3. #3Qwen2 5 Coder 7bAlibaba - 33K context$0.010/M
  4. #4gpt-oss-20bOpenAI - 131K context$0.015/M
  5. #5Llama3 2 11b VisionMeta - 131K context$0.015/M
  6. #6Llama3 2 3bMeta - 131K context$0.015/M
  7. #7Granite 4.0 MicroIbm Granite - 131K context$0.017/M
  8. #8Mistral NemoMistral AI - 131K context$0.019/M
  9. #9Llama 3.2 3B InstructMeta - 131K context$0.020/M
  10. #10Llama Guard 3 8BMeta - 131K context$0.020/M

Lowest output cost per 1M tokens

Useful for long-form generation, agents, code output, and other completion-heavy workloads.

  1. #1Flux 1 Dev Controlnet UnionFireworks Ai - 4K context$0.0010/M
  2. #2Llama 3.2 3B InstructMeta - 131K context$0.020/M
  3. #3Paddleocr VlNovita - 16K context$0.020/M
  4. #4Deepseek R1 Distill Llama 8bMeta - 131K context$0.025/M
  5. #5Llama3 2 11b VisionMeta - 131K context$0.025/M
  6. #6Llama3 2 3bMeta - 131K context$0.025/M
  7. #7Deepseek OcrDeepseek - 8K context$0.030/M
  8. #8Ling-2.6-flashInclusionai - 262K context$0.030/M
  9. #9Llama 3.1 8BMeta - 128K context$0.030/M
  10. #10Meta Llama 3 1 8b InstructMeta - 131K context$0.030/M

Start with your workload

Estimate the ratio of prompt tokens to generated tokens. Rank by the side that drives most of your bill, then calculate the combined cost.

Check capability and limits

A low token rate only helps if the model supports the required context, output length, modality, and tool-calling behavior.

Verify the final provider

List prices can exclude caching, batch discounts, minimum spend, and regional fees. Confirm the source shown on the model page before purchasing.

Estimate your token costsBrowse all modelsCompare two models