whatmodel.sh / image / FLUX.1 [schnell]
FLUX.1 [schnell]
↗ share
+ contribute
FLUX.1 [schnell] is an open-weights image model from Black Forest Labs, released August 2024. Its published size is 12B parameters. Licence: apache-2.0.
Score within the LMArena text-to-image comparison set only.
$ how it compares
The image models ranked either side of FLUX.1 [schnell], out of 197 with a score, then more from Black Forest Labs.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 133 |  BAGEL | 42±1 | api $0.10/image +local | ||
| 134 |  Midjourney V6.1 | 42±4 | comparison only | ||
| 135 | FLUX.1 [schnell] | 42±2 | api $0.0030/image +local | ||
| 136 |  OmniGen2 | 42±2 | api $0.15/image +local | ||
| 137 |  Stable Diffusion 3.5 Large | 42±1 | api $0.065/image +local | ||
| more from Black Forest Labs | |||||
| 32 | FLUX.2 [max] | 64±1 | api $0.070/image | ||
| 36 | FLUX.2 [pro] | 63±2 | api $0.030/image | ||
| 37 | FLUX.2 [flex] | 62±2 | api $0.050/image | ||
$ what it's good at
text to image▂▅▅▃▄▄█▃▁▂▁▁ #83 of 126 * · 59% of its score · 3 tests
Beats Latent Diffusion text2img-large (LDM-KL-8-G, LAION-400M), VQ-Diffusion (ITHQ), FLUX.1 Kontext [dev] here, though they rank higher overall. Trails Playground v2 (1024px aesthetic) here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| T2I-ReasonBench (overall) | 57.6% ✎ | ▅▂█▃▂▂····▂▂ beats 9/13 | 48% |
| WISE (WiScore overall, GPT-4o judge) | 40.0% ✎ | ▄·▆▂▆█▄▆▂▄·▄ beats 7/22 | 11% |
| GenEval | 67.7% ✎ | ▇▃▅▇▁▄▅▃█▃▃▅ beats 18/39 | — |
other tests2 not in a task · 39% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Rapidata human-face generation alignment (Elo) | 885.1 ✎ | ▃··▅▃··▃▃··█ beats 0/8 | 21% |
| VVRBench accuracy (10,000 verifiable synthetic-scene tasks) | 2.9% ✎ | ▃·█▅▃·▅▃···▃ beats 3/10 | 18% |
$ where to run it
Through an API
api · Replicate · $0.0030/image · replicate.com ↗ · price sourceUSD per output image. Text-to-image; one output image, standard FLUX.1 schnell endpoint. Replicate output/character billing; hardware-time rates are not added to this tariff. Taxes and separate services excluded. Bertrande, https://bertrande.com/en/ai/replicate/black-forest-labs--flux-schnell; observed 2026-09-20T01:44:15Z. CC BY 4.0; units normalized and model identifier mapped.
Other ways to run it
version
- Run it as published · needs ~7.7 GB, Mac only · from modelfit.io
uvx --from mflux mflux-generate --model schnell --prompt "a cat holding a sign" --steps 2 -q 4
Official weights and documentation: huggingface.co/black-forest-labs ↗
$ how the score was worked out · T2I-ReasonBench: Benchmarking… 48%, Rapidata/Face_Generation_Benc… 21%, Verifiable Visual Rewards Tra… 18% and 1 more
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 |
|---|---|---|
 T2I-ReasonBench: Benchmarking Reasoning-Informed Text-to-Image Generation (Findings of ACL 2026) | 1 | 48% |
 Rapidata/Face_Generation_Benchmark (Human Face Generation Alignment) published dataset-card leaderboard | 1 | 21% |
 Verifiable Visual Rewards Transfer from Synthetic Scenes to Natural Prompts (VVRBench), arXiv:2609.35641v1 | 1 | 18% |
| WISE: A World Knowledge-Informed Semantic Evaluation for Text-to-Image Generation (arXiv v4) | 1 | 11% |
| prior · pull toward the middle | 2% | |
| weighted average | = 41.6 |
not counted (1)
| Breaking the Accuracy-Speed Barrier: How MXFP4/6 Quantization Revolutionizes Image and Video Generation | GenEval [correct] | 67.7% | fewer than three models have this test yet, or this source ran it for this model alone |
$ 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