whatmodel.sh / image / Photon
Photon
↗ share
+ contribute
Photon is a closed-weights image model from Luma AI.
Score within the LMArena text-to-image comparison set only.
$ how it compares
The image models ranked either side of Photon, out of 197 with a score, then more from Luma AI.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 97 |  FLUX.2 [klein] 4B | 52±2 | api $0.0011/image +local | ||
| 98 |  P-Image | 51±2 | api $0.0050/image | ||
| 99 | Photon | 51±2 | api $0.019/image | ||
| 100 |  Recraft V3 | 50±2 | api $0.040/image | ||
| 101 | FLUX.1 Kontext [pro] | 50±2 | api $0.040/image | ||
| more from Luma AI | |||||
| 21 | Uni-1.1 | 67±2 | api $0.040/image | ||
| 27 | Uni-1.1 Max | 65±2 | api $0.10/image | ||
$ what it's good at
text to image▂▅▅▃▄▄█▃▁▂▁▁ #66 of 126 · 84% of its score · 3 tests
Beats FLUX.2 [klein] 4B, P-Image here, though they rank higher overall. Trails FLUX.1 Kontext [pro] here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| LMArena text-to-image | 1035.7 ✎ | ▂▂▃▄▇█▄▂▄▂·▂ beats 14/75 | 56% |
| GenEval | 73.8% ✎ | ██·▄·▄·█···▄ beats 7/8 | 18% |
| MMMG image generation (average over 15 tasks) | 58.7% ✎ | █··█▄▄·▄▄·▄▄ beats 6/9 | 10% |
other tests1 not in a task · 14% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Rapidata human-face generation alignment (Elo) | 955.5 ✎ | ▃··▅▃··▃▃··█ beats 1/8 | 14% |
$ where to run it
Through an API
api · fal.ai · $0.019/image · fal.ai ↗ · price sourceUSD for one 1-megapixel output: fal bills fal-ai/luma-photon at $0.019 per output megapixel (price = rate × 1 MP; fal's rounding for 1024×1024 ≈ 1.05 MP is not stated; larger outputs cost proportionally more). Product: fal.ai partner API (Luma's model served through fal). API model: fal-ai/luma-photon (Luma Photon, not Photon Flash). Settings: text-to-image, 1 MP output, one image. Price basis: published per-megapixel list rate × 1 MP. The fal endpoint is public and not marked deprecated; Luma's own current Agents API lists only uni-1, uni-1-max and ray-3.2, so this is not a Luma direct tariff. Currency: USD. fal.ai official model page, observed 2026-09-26.
$ how the score was worked out · LMArena leaderboard dataset 56%, MMMG: a Comprehensive and Rel… 28%, Rapidata/Face_Generation_Benc… 14%
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 | 56% |
 MMMG: a Comprehensive and Reliable Benchmark for Multitask Multimodal Generation (arXiv:2505.17613v2) | 2 | 28% |
 Rapidata/Face_Generation_Benchmark (Human Face Generation Alignment) published dataset-card leaderboard | 1 | 14% |
| prior · pull toward the middle | 2% | |
| weighted average | = 50.6 |
$ 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