by OpenAIdeprecated
Deprecated. OpenAI's first multimodal image gen+edit model — successor to DALL-E 3, predecessor to GPT Image 2. Higher per-token cost than GPT Image 2 with slower speed; new work should use GPT Image 2.
GPT Image 1 (gpt-image-1) was OpenAI's first natively multimodal image
generation + editing model — successor to DALL-E 3, predecessor to GPT Image 2.
Currently DEPRECATED in favor of GPT Image 2.
Deprecated — still callable via API but no longer recommended for new
work. Migration path: switch apiId to gpt-image-2 for higher fidelity
and better edit support, or drop to dall-e-3 for the cheap-and-classic
text-rendering use case.
Unsupported: audio, video, streaming, function calling, structured outputs, fine-tuning.
POST /v1/images/generations — text-to-imagePOST /v1/images/edits — inpaint / editToken-based:
| Modality | Input (per 1M) | Cached input | Output (per 1M) |
|---|---|---|---|
| Image | $10.00 | $2.50 | $40.00 |
| Text | $5.00 | $1.25 | — |
Per-image (1024×1024):
| Quality | Price |
|---|---|
| Low | $0.011 |
| Medium | $0.042 |
| High | $0.167 |
Roughly 25% more expensive on image-output tokens than GPT Image 2 — one of the reasons it was retired.
Almost never for new code — pick gpt-image-2 instead. The only reason to
keep gpt-image-1 active is API compatibility for an existing
production integration that hasn't been re-tested against gpt-image-2.
Quick tips from the community about what works with GPT Image 1 right now.
Sign in to share a tip.
No tips yet. Add a tip for this model.