Spaces:

Duplicated from zerogpu-aoti/FLUX.1-dev-fa3-aoti

zerogpu-aoti
/

FLUX.1-dev-fa3-aoti-blocks-load

Runtime error

App Files Files Community

FLUX.1-dev-fa3-aoti-blocks-load

Commit History

Update README.md

2dda6d2
verified

cbensimon HF Staff commited on Sep 12, 2025

Update app.py

7359b93
verified

cbensimon HF Staff commited on Sep 10, 2025

Update aoti.py

96ac21f
verified

cbensimon HF Staff commited on Sep 10, 2025

Update app.py

eda2d0c
verified

cbensimon HF Staff commited on Sep 10, 2025

AOTI load

4e94151

cbensimon HF Staff commited on Sep 10, 2025

Update optimization.py

39286c5
verified

cbensimon HF Staff commited on Sep 5, 2025

Update optimization.py

c628e2d
verified

cbensimon HF Staff commited on Sep 5, 2025

Update optimization.py

38cc643
verified

cbensimon HF Staff commited on Sep 5, 2025

Regional AoT

780a177
verified

cbensimon HF Staff commited on Sep 5, 2025

Update README.md

d6785d3
verified

cbensimon HF Staff commited on Aug 29, 2025

bf16 fa3

c60d44a

cbensimon HF Staff commited on Aug 29, 2025

Enable aoti

7301ed0

cbensimon HF Staff commited on Aug 29, 2025

fp8e4m3 (disable aoti)

c72054d

cbensimon HF Staff commited on Aug 29, 2025

gr.Gallery()

48a26b4

cbensimon HF Staff commited on Aug 29, 2025

FLUX.1-dev

bf738a2

cbensimon HF Staff commited on Aug 29, 2025

fuse_qkv_projections

a5c23d1

cbensimon HF Staff commited on Aug 29, 2025

Revert "qkv"

da26724

cbensimon HF Staff commited on Aug 29, 2025

qkv

cfad87d

cbensimon HF Staff commited on Aug 29, 2025

fa3 + big refresh

69667cb

cbensimon HF Staff commited on Aug 29, 2025

Demo touch-ups

da628cb

cbensimon HF Staff commited on Jul 3, 2025

No progress for now

856d03c

cbensimon HF Staff commited on Jul 3, 2025

Empty commit

9e76b53

cbensimon HF Staff commited on Jul 3, 2025

9 images + No pipeline progress

bb62642

cbensimon HF Staff commited on Jul 3, 2025

Useless here

8b98990

cbensimon HF Staff commited on Jul 3, 2025

Streaming

f742b99

cbensimon HF Staff commited on Jul 3, 2025

Touch ups

24ec8bb

cbensimon HF Staff commited on Jul 3, 2025

Further optimize + quantize

ce8b907

cbensimon HF Staff commited on Jul 3, 2025

No need torchvision

49c2af2

cbensimon HF Staff commited on Jul 3, 2025

Not needed anymore

7c48676

cbensimon HF Staff commited on Jul 3, 2025

Fixes

747ebd1

cbensimon HF Staff commited on Jul 3, 2025

PyTorch 2.9 issue?

a061af5

cbensimon HF Staff commited on Jul 3, 2025

Real file

879645c

cbensimon HF Staff commited on Jul 3, 2025

Cleanup

f69d27d

cbensimon HF Staff commited on Jul 3, 2025

Alright

d41da38

cbensimon HF Staff commited on Jul 3, 2025

ZeroGPUCompiledModel

674e245

cbensimon HF Staff commited on Jul 3, 2025

Timings

a3d55a6

cbensimon HF Staff commited on Jul 2, 2025

Restore transformer config

c29a475

cbensimon HF Staff commited on Jul 1, 2025

Duration

50d5e76

cbensimon HF Staff commited on Jul 1, 2025

Some basic requirements

984c247

cbensimon HF Staff commited on Jul 1, 2025

Begin actual demo

9544e60

cbensimon HF Staff commited on Jul 1, 2025

Brainstorming backup

4d0c9f3

cbensimon HF Staff commited on Jun 27, 2025

Two calls in a row

04ce204

cbensimon HF Staff commited on Jun 27, 2025

Worker re-use

4f27510

cbensimon HF Staff commited on Jun 27, 2025

Move weights to CUDA

060b6b6

cbensimon HF Staff commited on Jun 27, 2025

Fix CUDA path

66ca860

cbensimon HF Staff commited on Jun 27, 2025

Package AOTI

495d59a

cbensimon HF Staff commited on Jun 27, 2025

detach after copy_

9292d1e

cbensimon HF Staff commited on Jun 27, 2025

Keep pinned-memory

f1f06d3

cbensimon HF Staff commited on Jun 27, 2025

requires_grad=False

4166007

cbensimon HF Staff commited on Jun 27, 2025

Load constants

8012ef2

cbensimon HF Staff commited on Jun 27, 2025