Commits · MySafeCode/AnyTalker

Update GPU duration to 600 seconds (10 minutes)

60f55fe

C4G-HKUST commited on 28 days ago

Update GPU duration to 1800 seconds (30 minutes)

98ffd05

C4G-HKUST commited on 28 days ago

Update generation time estimates and GPU duration

26fd4c9

C4G-HKUST commited on 28 days ago

Fix examples: move generation time to second column and fix list index out of range error

e9633ab

C4G-HKUST commited on 28 days ago

fix: hf demo

b505615

C4G-HKUST commited on 28 days ago

Fix wav2vec model path to use Hugging Face model ID: facebook/wav2vec2-base-960h

b2737a3

C4G-HKUST commited on 28 days ago

Fix VAE scale tensor device mismatch: recreate scale list on GPU and move T5 encoder

c2db27b

C4G-HKUST commited on 29 days ago

Move models from CPU to GPU in worker process (like LivePortrait)

f2436d3

C4G-HKUST commited on 29 days ago

Simplify GPU handling following LivePortrait pattern: CPU in main process, GPU in worker via @spaces.GPU

7b63b72

C4G-HKUST commited on 29 days ago

Fix Stateless GPU environment: avoid CUDA init in main process, initialize GPU in worker process

f0617b0

C4G-HKUST commited on 29 days ago

Track GPU device during initialization and use same GPU in worker process and inference

3c322ce

C4G-HKUST commited on 29 days ago

Add GPU availability check in worker process for @spaces.GPU decorator

694fec8

C4G-HKUST commited on 29 days ago

Fix Gradio version and launch config: downgrade to 5.0.x, remove show_api parameter

5cb609b

C4G-HKUST commited on 29 days ago

Update: increase GPU duration to 360s, add gradio==5.42.0, update README

7569450

C4G-HKUST commited on 29 days ago

Use @spaces.GPU decorator for GPU management (reference LivePortrait)

f1ae228

C4G-HKUST commited on 29 days ago

Fix GPU detection and CPU fallback: remove ssr_mode, add runtime GPU check, support CPU mode for FaceInference

b7867cd

C4G-HKUST commited on 29 days ago

Fix CUDA device fallback in WanVAE and WanAF2V

7c629b9

C4G-HKUST commited on 29 days ago

Fix CUDA device access in T5EncoderModel and add GPU check

831eccf

C4G-HKUST commited on 29 days ago

Fix Gradio launch config for HF Spaces (remove no_proxy, use queue)

ef30698

C4G-HKUST commited on 29 days ago

Use flash_attn 2.7.4.post1 precompiled wheel (torch 2.6 compatible)

28f2de3

zzz66 commited on 29 days ago

Remove flash_attn (no precompiled wheel for torch 2.6)

bd2c73d

zzz66 commited on 29 days ago

Fix flash_attn wheel URL (cu122)

e7afc0e

zzz66 commited on 29 days ago

Use precompiled flash_attn wheel

96465d6

zzz66 commited on 29 days ago

Use preload_from_hub for models, only upload vocal_separator

63a1b7e

zzz66 commited on 29 days ago

Add YAML frontmatter for HF Space config

da470b8

zzz66 commited on 29 days ago

Initial commit with LFS

400a879

zzz66 commited on 29 days ago

initial commit

d5bbcc5
verified

zzz66 commited on 29 days ago

Spaces:

Duplicated from C4G-HKUST/AnyTalker

MySafeCode
/

AnyTalker

Sleeping

Commit History

Update GPU duration to 600 seconds (10 minutes)

60f55fe

Update GPU duration to 1800 seconds (30 minutes)

98ffd05

Update generation time estimates and GPU duration

26fd4c9

Fix examples: move generation time to second column and fix list index out of range error

e9633ab

fix: hf demo

b505615

Fix wav2vec model path to use Hugging Face model ID: facebook/wav2vec2-base-960h

b2737a3

Fix VAE scale tensor device mismatch: recreate scale list on GPU and move T5 encoder

c2db27b

Move models from CPU to GPU in worker process (like LivePortrait)

f2436d3

Simplify GPU handling following LivePortrait pattern: CPU in main process, GPU in worker via @spaces.GPU

7b63b72

Fix Stateless GPU environment: avoid CUDA init in main process, initialize GPU in worker process

f0617b0

Track GPU device during initialization and use same GPU in worker process and inference

3c322ce

Add GPU availability check in worker process for @spaces.GPU decorator

694fec8

Fix Gradio version and launch config: downgrade to 5.0.x, remove show_api parameter

5cb609b

Update: increase GPU duration to 360s, add gradio==5.42.0, update README

7569450

Use @spaces.GPU decorator for GPU management (reference LivePortrait)

f1ae228

Fix GPU detection and CPU fallback: remove ssr_mode, add runtime GPU check, support CPU mode for FaceInference

b7867cd

Fix CUDA device fallback in WanVAE and WanAF2V

7c629b9

Fix CUDA device access in T5EncoderModel and add GPU check

831eccf

Fix Gradio launch config for HF Spaces (remove no_proxy, use queue)

ef30698

Use flash_attn 2.7.4.post1 precompiled wheel (torch 2.6 compatible)

28f2de3

Remove flash_attn (no precompiled wheel for torch 2.6)

bd2c73d

Fix flash_attn wheel URL (cu122)

e7afc0e

Use precompiled flash_attn wheel

96465d6

Use preload_from_hub for models, only upload vocal_separator

63a1b7e

Add YAML frontmatter for HF Space config

da470b8

Initial commit with LFS

400a879

initial commit

d5bbcc5
verified

Commit History

Update GPU duration to 600 seconds (10 minutes) 60f55fe

Update GPU duration to 1800 seconds (30 minutes) 98ffd05

Update generation time estimates and GPU duration 26fd4c9

Fix examples: move generation time to second column and fix list index out of range error e9633ab

fix: hf demo b505615

Fix wav2vec model path to use Hugging Face model ID: facebook/wav2vec2-base-960h b2737a3

Fix VAE scale tensor device mismatch: recreate scale list on GPU and move T5 encoder c2db27b

Move models from CPU to GPU in worker process (like LivePortrait) f2436d3

Simplify GPU handling following LivePortrait pattern: CPU in main process, GPU in worker via @spaces.GPU 7b63b72

Fix Stateless GPU environment: avoid CUDA init in main process, initialize GPU in worker process f0617b0

Track GPU device during initialization and use same GPU in worker process and inference 3c322ce

Add GPU availability check in worker process for @spaces.GPU decorator 694fec8

Fix Gradio version and launch config: downgrade to 5.0.x, remove show_api parameter 5cb609b

Update: increase GPU duration to 360s, add gradio==5.42.0, update README 7569450

Use @spaces.GPU decorator for GPU management (reference LivePortrait) f1ae228

Fix GPU detection and CPU fallback: remove ssr_mode, add runtime GPU check, support CPU mode for FaceInference b7867cd

Fix CUDA device fallback in WanVAE and WanAF2V 7c629b9

Fix CUDA device access in T5EncoderModel and add GPU check 831eccf

Fix Gradio launch config for HF Spaces (remove no_proxy, use queue) ef30698

Use flash_attn 2.7.4.post1 precompiled wheel (torch 2.6 compatible) 28f2de3

Remove flash_attn (no precompiled wheel for torch 2.6) bd2c73d

Fix flash_attn wheel URL (cu122) e7afc0e

Use precompiled flash_attn wheel 96465d6

Use preload_from_hub for models, only upload vocal_separator 63a1b7e

Add YAML frontmatter for HF Space config da470b8

Initial commit with LFS 400a879

initial commit d5bbcc5 verified

Update GPU duration to 600 seconds (10 minutes)

60f55fe

Update GPU duration to 1800 seconds (30 minutes)

98ffd05

Update generation time estimates and GPU duration

26fd4c9

Fix examples: move generation time to second column and fix list index out of range error

e9633ab

fix: hf demo

b505615

Fix wav2vec model path to use Hugging Face model ID: facebook/wav2vec2-base-960h

b2737a3

Fix VAE scale tensor device mismatch: recreate scale list on GPU and move T5 encoder

c2db27b

Move models from CPU to GPU in worker process (like LivePortrait)

f2436d3

Simplify GPU handling following LivePortrait pattern: CPU in main process, GPU in worker via @spaces.GPU

7b63b72

Fix Stateless GPU environment: avoid CUDA init in main process, initialize GPU in worker process

f0617b0

Track GPU device during initialization and use same GPU in worker process and inference

3c322ce

Add GPU availability check in worker process for @spaces.GPU decorator

694fec8

Fix Gradio version and launch config: downgrade to 5.0.x, remove show_api parameter

5cb609b

Update: increase GPU duration to 360s, add gradio==5.42.0, update README

7569450

Use @spaces.GPU decorator for GPU management (reference LivePortrait)

f1ae228

Fix GPU detection and CPU fallback: remove ssr_mode, add runtime GPU check, support CPU mode for FaceInference

b7867cd

Fix CUDA device fallback in WanVAE and WanAF2V

7c629b9

Fix CUDA device access in T5EncoderModel and add GPU check

831eccf

Fix Gradio launch config for HF Spaces (remove no_proxy, use queue)

ef30698

Use flash_attn 2.7.4.post1 precompiled wheel (torch 2.6 compatible)

28f2de3

Remove flash_attn (no precompiled wheel for torch 2.6)

bd2c73d

Fix flash_attn wheel URL (cu122)

e7afc0e

Use precompiled flash_attn wheel

96465d6

Use preload_from_hub for models, only upload vocal_separator

63a1b7e

Add YAML frontmatter for HF Space config

da470b8

Initial commit with LFS

400a879

initial commit

d5bbcc5
verified