Models that return images
These return pictures, and pictures are billed as image output tokens — one image costs several times what a paragraph of text costs. Check your balance before a bulk run: the hold placed on your wallet scales with the number of images you ask for.
10 models in this category · prices in toman per 1M tokens
| Model | Capabilities | Context | Input / 1M | Output / 1M | ≈ 1,000 words |
|---|---|---|---|---|---|
 Nano Banana 2 Lite (Gemini 3.1 Flash Lite Image)google/gemini-3.1-flash-lite-image | Image inputReasoningJSON outputImage output | 65,536 | 86,555 | 519,329 | 788 |
Nano Banana (Gemini 2.5 Flash Image)google/gemini-2.5-flash-image | Image inputReasoningJSON outputImage output | 32,768 | 103,866 | 865,548 | 1,260 |
Nano Banana 2 (Gemini 3.1 Flash Image)google/gemini-3.1-flash-image | Image inputReasoningJSON outputImage output | 131,072 | 173,110 | 1,038,658 | 1,575 |
Nano Banana 2 (Gemini 3.1 Flash Image Preview)google/gemini-3.1-flash-image-preview | Image inputReasoningJSON outputImage output | 65,536 | 173,110 | 1,038,658 | 1,575 |
GPT-5 Image Miniopenai/gpt-5-image-mini | Image inputReasoningJSON outputFile input | 400,000 | 865,548 | 692,439 | 2,025 |
Nano Banana 2.1google/gemini-nano-banana-2.1 | Image inputTool callingReasoningJSON output | 65,536 | 519,329 | 2,596,645 | 4,051 |
Nano Banana Pro (Gemini 3 Pro Image)google/gemini-3-pro-image | Image inputTool callingReasoningJSON output | 131,072 | 692,439 | 4,154,632 | 6,301 |
Nano Banana Pro (Gemini 3 Pro Image Preview)google/gemini-3-pro-image-preview | Image inputReasoningJSON outputImage output | 65,536 | 692,439 | 4,154,632 | 6,301 |
GPT-5 Imageopenai/gpt-5-image | Image inputReasoningJSON outputFile input | 400,000 | 3,462,194 | 3,462,194 | 9,002 |
GPT-5.4 Image 2openai/gpt-5.4-image-2 | Image inputReasoningJSON outputFile input | 272,000 | 2,769,755 | 5,193,290 | 10,352 |
Other categories
Models that hold up in a real codebaseCheap models for high-volume workFree models, shared capacityModels that read imagesReasoning models, and what they costLong-context models for documents and codebasesModels that call your tools reliablyEmbedding models for semantic searchBatch variants (cheaper, slower)