POST

/v1/estimate · fal-ai/trellis-2-lora-trainer

field	recorded source
pricing	Source linked above. Source trace `037c76e3ff92c4422109541f0da76bebb1f6229fc3b2181419aef5c0f6c8f1f1`

num_objects number · usage
reuse_prepared_dataset boolean · usage
denoiser enum · usage
steps number · usage
rank enum · usage
resolution enum · usage
quote	$9.3423 — num_objects 10 · reuse_prepared_dataset false · denoiser sparse_structure · steps 1000 · rank 32 · resolution 512
examples	$9.3423 — one-time dataset-preparation charge of $0.10023 per input object (10-object minimum $8.34 — waived when you reuse a prepared dataset $9.3423 — $8.34 per 1,000 steps for sparse_structure $4.2523 — $3.25 per 1,000 steps for geometry $4.3823 — $3.38 per 1,000 steps for texture $5.1723 — scaled by your rank (×0.5 for 16 $17.6823 — ×2 for 64 $17.6823 — resolution (×2 for 1024