[model] support FLUX.1-Kontext-dev#561
Conversation
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
vllm_omni/diffusion/models/flux_kontext/flux_kontext_transformer.py
Outdated
Show resolved
Hide resolved
RuixiangMa
changed the title
Signed-off-by: Lancer <maruxiiang6688@gmail.com>
|
@Bounty-hunter @wtomin PTAL |
RuixiangMa
force-pushed
the
flux1
branch
2 times, most recently
from
65f308f to
b260332
Compare
Signed-off-by: Lancer <maruixiang6688@gmail.com>
RuixiangMa
force-pushed
the
flux1
branch
3 times, most recently
from
503c037 to
e9b6432
Compare
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: e9b6432d70
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
Signed-off-by: Lancer <maruixiang6688@gmail.com>
Signed-off-by: Lancer <maruixiang6688@gmail.com>
|
if you use offloading, please also share your latency here with a table |
|
done |
Signed-off-by: Lancer <maruixiang6688@gmail.com>
|
Refactored flux_kontext，move components into the flux folder |
2 participants
Purpose
FLUX.1-Kontext-dev model support
Test Plan
Test Result
4 * NVIDIA 4090(24G)
vllm serve black-forest-labs/FLUX.1-Kontext-dev --omni --port 8004 --enable_cpu_offload --tensor_parallel_size 4curl -s -X POST "http://localhost:8004/v1/images/edits" -F "image=@test.jpg" -F "prompt=Change the sky to orange sunset." -F "guidance_scale=1.0" -F "num_inference_steps=50" -F "n=1" -F "size=1024x1024" -F "output_format=png" | jq -r '.data[0].b64_json' | base64 --decode > output.png