foundationvision/infinity 🔢📝 → 🖼️

▶️ 431 runs 📅 Dec 2024 ⚙️ Cog 0.9.23 🔗 GitHub 📄 Paper ⚖️ License

autoregressive text-to-image

About

Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesis

Example Output

Prompt:

"alien spaceship enterprise"

Output

Performance Metrics

5.26s Prediction Time

162.03s Total Time

All Input Parameters

{
  "tau": 0.5,
  "prompt": "alien spaceship enterprise",
  "guidance_scale": 3
}

Input Parameters

tau Type: numberDefault: 0.5: tau in self attention
seed Type: integer: Random seed. Leave blank to randomize the seed
prompt Type: stringDefault: alien spaceship enterprise: Input prompt
guidance_scale Type: numberDefault: 3Range: 1 - 10: Scale for classifier-free guidance

Output Schema

Output

Type: string • Format: uri

Example Execution Logs

Using seed: 2533
prompt=alien spaceship enterprise
cfg: [3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0, 3.0], tau: [0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5]
/src/tools/run_infinity.py:112: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
with torch.cuda.amp.autocast(enabled=True, dtype=torch.bfloat16, cache_enabled=True):
/src/infinity/models/basic.py:495: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
with torch.cuda.amp.autocast(enabled=False):    # disable half precision
/src/infinity/models/basic.py:495: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
with torch.cuda.amp.autocast(enabled=False):    # disable half precision
/src/infinity/models/basic.py:495: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
with torch.cuda.amp.autocast(enabled=False):    # disable half precision
/src/infinity/models/basic.py:495: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
with torch.cuda.amp.autocast(enabled=False):    # disable half precision
/src/infinity/models/basic.py:495: FutureWarning: `torch.cuda.amp.autocast(args...)` is deprecated. Please use `torch.amp.autocast('cuda', args...)` instead.
with torch.cuda.amp.autocast(enabled=False):    # disable half precision
cost: 5.212285041809082, infinity cost=4.984928131103516

Version Details

Version ID: fd89a074acb4ea68dfa14eac11eb079d9a5dfaf063d70f7e2afc465c9c62af60
Version Created: December 24, 2024

Run on Replicate →