Query Standards for Merlin
Welcome to Merlin AI, where we strive to provide you with the best AI purchase you ever make. Below, you will find a breakdown of the query ratios for each AI model available through Merlin
AI Model Queries deduction per use
- GPT 4o 15
- GPT 4o Mini1
- GPT 4o Mini (Long context)5
- GPT 3.51
- Gemini 1.5 Pro15
- Gemini 1.5 Flash1
- Gemini 1.5 Flash (Long context)5
- Claude 3 Opus50
- Claude 3.5 Sonnet25
- Claude 3 Sonnet10
- Claude 3 Haiku1
- Claude 3 Haiku (Long context)5
- Mistral Large25
- Mistral1
- Llama 3.1 405B15
- Flux 1.1 Pro100
- Flux 1 Pro140
- Flux Schnell10
- Dalle 3 / Render Works v3100
When live search mode or web-access is enabled, 2x queries are deducted from whats mentioned above as the overall context sent to LLMs is increase and search APIs have extra associated costs.
Merlin magic auto-selects the model and mode for you based on your query, and queries are cut based on which model and mode was selected, pricing outlined above.
For features like Youtube summarizer, Blog Summarizer, webpage summariser, etc - the queries depends on usage required (such as length, subtitles and language of videos in case of youtube). Feel free to explore the capabilities of each AI model and optimize your queries accordingly.
Should you have any further questions or require assistance, do not hesitate to reach out to our support team at support@getmerlin.in. Note that Merlin team re-adjusts query mapping on a dynamic basis to accomodate for actual costs across different models and usecases. And therefore this page is dynamically updated