Verified: 4 months ago
SafeTensor
The FLUX.1 [dev] Model is licensed by Black Forest Labs. Inc. under the FLUX.1 [dev] Non-Commercial License. Copyright Black Forest Labs. Inc.
IN NO EVENT SHALL BLACK FOREST LABS, INC. BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.
Mini list aircraft sdxlV1
This is a version for SDXL, it knows quite a lot of aircraft, and to get a result it's enough to specify the trigger word. It produces a somewhat noisy output — I probably overdid it with the learning rate.
Additionally, you can also include words like dogfight, bomb, aircraft, 2aircraft, many_aircraft.
Aviation LoRA Flux D V0.5
This experimental LoRA can generate images of several aircraft types. It was trained on my personal datasets, combined into one large model. After tests, I realise the datasets will need to be rebuilt for a full-scale LoRA, so consider this release strictly experimental.
Use at your own risk: my single-aircraft LoRAs still produce more reliable results.
Supported aircraft (trigger words) and estimated accuracy
Wright_Flyer - low accuracy, ≈ 1 in 8 generations successful
vickers_vimy aircraft - low accuracy, ≈ 1 in 8 generations successful
Fokker_Dr.I aircraft - acceptable quality
Douglas_DC-3 - acceptable quality
B_17 bomber - medium accuracy, about half of the images look correct
P-47_Thunderbolt - acceptable quality
Messerschmitt_Me_262 - low accuracy, ≈ 1 in 12 generations successful
F-86_Sabre - medium accuracy, about 50 % correct
MIG_15 - medium accuracy, about 50 % correct
F-4_Phantom_II - acceptable quality
Boeing_707 - medium accuracy, about 50 % correct
Concorde - acceptable quality
F-117 - low accuracy, ≈ 1 in 12 generations successful
SR-71_Blackbird - low accuracy, ≈ 1 in 12 generations successful