whatmodel.sh / image / ChatGPT Image Latest (chatgpt-image-latest)
ChatGPT Image Latest (chatgpt-image-latest)
↗ share
+ contribute
ChatGPT Image Latest (chatgpt-image-latest) is a closed-weights image model from OpenAI, released December 2025. Licence: proprietary.
Catalogue identity notes
OpenAI API model chatgpt-image-latest, released on 16 Dec 2025 alongside gpt-image-1.5 (OpenAI API changelog: 'Released gpt-image-1.5 and chatgpt-image-latest'). Its model page now reads 'Previous image model used in ChatGPT ... This points to the Image snapshot previously used in ChatGPT' and lists the single alias chatgpt-image-latest. Arena image_edit entry 'chatgpt-image-latest-high-fidelity (20251216)' (openai, Proprietary; listed since 2026-01-26) carries that release date and the high-fidelity edit setting. Kept separate from gpt-image-1-5: OpenAI lists them as separate models and Arena rates them as separate entries (1388.6 vs 1369.7 for 'gpt-image-1.5-high-fidelity'). Caveat: OpenAI publishes no dated snapshot for the alias. No text_to_image Arena entry.
Model information sources
Score within the LMArena text-to-image comparison set only.
$ how it compares
The image models ranked either side of ChatGPT Image Latest (chatgpt-image-latest), out of 197 with a score, then more from OpenAI.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 12 |  Seedream 5.0 Pro | 75±2 | api $0.050/image | ||
| 13 |  Reve 2.0 | 74±2 | api $0.25/image | ||
| 14 | ChatGPT Image Latest (chatgpt-image-latest) | 74±4 | api $0.034/image | ||
| 15 |  MAI-Image-2.5 | 74±2 | comparison only | ||
| 16 |  Nano Banana 2 Lite (Gemini 3.1 Flash-Lite Image) | 73±2 | api $0.034/image | ||
| more from OpenAI | |||||
| 1 | GPT Image 2.5 Sunburst | 92±2 | api $0.013/image | ||
| 2 | GPT Image 2.5 Flare | 88±2 | api $0.013/image | ||
| 3 | GPT Image 2 | 85±1 | api $0.053/image | ||
$ what it's good at
image to image / editing▁▁·▃▂▃█▃▃▄▁▁ #9 of 62 * · 93% of its score · 1 test
Beats Reve 2.1, Nano Banana 2 (Gemini 3.1 Flash Image), Reve 2.0 here, though they rank higher overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Arena image edit | 1388.6 ✎ | ▂▁▁▂▆▇█▄▆▅▁▂ beats 44/53 | 93% |
$ where to run it
Through an API
api · $0.034/image · developers.openai.com ↗ · price sourceUSD for one generated 1024x1024 medium-quality image on the OpenAI API: the model page's Image generation tables list 1024x1024 at Low $0.009, Medium $0.034, High $0.133 per image (1024x1536/1536x1024: $0.013 / $0.05 / $0.2). Text input ($5 per 1M) and image input ($8 per 1M) tokens are additional. Product: OpenAI Image API (v1/images/generations and v1/images/edits; Batch supported). API model: chatgpt-image-latest (only alias listed; no dated snapshot). Settings: quality=medium, 1024x1024, n=1. Price basis: published per-image list rate. Currency: USD. OpenAI official model page, observed 2026-10-09.
$ how the score was worked out · LMArena leaderboard dataset 93%
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 LMArena leaderboard dataset | 1 | 93% |
| prior · pull toward the middle | 7% | |
| weighted average | = 74.2 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json