whatmodel.sh / image / GPT-Image-2.6 Flare
GPT-Image-2.6 Flare
↗ share
+ contribute
GPT-Image-2.6 Flare is a closed-weights image model from OpenAI, released November 2026. Licence: proprietary.
Catalogue identity notes
OpenAI API changelog, Nov 4 2026 (released with Sunburst). Model id gpt-image-2.6-flare ('Image generation and editing for fast iteration'), single snapshot, generally available. Distinct from gpt-image-2-5-flare. Arena '(Medium)' = quality setting. Corresponds to Arena entries 'GPT-Image-2.6 Flare (Medium)' on text_to_image and image_edit. Price: No per-image price recorded; same reasons as gpt-image-2-6-sunburst. OpenAI token prices for gpt-image-2.6-flare: image input $8.00, cached image input $2.00, image output $30.00, text input $5.00, cached text $1.25 per 1M tokens (model page and pricing table). Observed 2026-11-12.
Model information sources
Score within the LMArena text-to-image comparison set only.
$ how it compares
The image models ranked either side of GPT-Image-2.6 Flare, out of 200 with a score, then more from OpenAI.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 2 | 87±2 | api $0.013/image | |||
| 3 | 87±2 | api price unknown | |||
| 4 | 85±2 | api price unknown | |||
| 5 | 83±1 | api $0.053/image | |||
| 6 | 80±2 | comparison only | |||
| more from OpenAI | |||||
| 1 | 91±2 | api $0.013/image | |||
| 14 | 73±4 | api $0.034/image | |||
| 19 | 70±1 | api $0.034/image | |||
$ what it's good at
text to image▂▅▅▃▄▄█▃▂▁▁▃ #4 of 129 * · 73% of its score · 1 test
| test | result | vs the others in its run | counts |
|---|---|---|---|
| LMArena text-to-image | 1383.5 ✎ | ▁▁▃▃▆█▄▂▂▂▃▁ beats 75/78 | 73% |
image to image / editing▁▁·▃▂▃█▆▃▆▃▂ #5 of 65 * · 25% of its score · 1 test
Trails GPT Image 2 here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Arena image edit | 1460.2 ✎ | ▂▁▁▃▆█▅▇█▄▃▃ beats 52/56 | 25% |
$ where to run it
Through an API
api · price unknown · developers.openai.com ↗
$ how the score was worked out · LMArena leaderboard dataset 98%
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 |
|---|---|---|
| 2 | 98% | |
| prior · pull toward the middle | 2% | |
| weighted average | = 84.5 |
$ 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