mike ce7b0e52e5 aa

2026-06-25 19:05:21 +02:00

.trash

aa

2026-06-21 04:21:40 +02:00

a6000-comfy

dphn/Dolphin3.0-Mistral-24B is the ungated mirror of the Dolphin 3.0 Mistral 24B — exactly what you asked for. It's ~48GB fp16, which needs GPU+CPU split (device_map="auto" with 32GB on GPU, ~16GB in RAM). Let me kick off the download and update the service in parallel.

2026-06-24 21:27:11 +02:00

gfpgan/weights

aa

2026-06-21 04:21:40 +02:00

tour-comfy

aa

2026-06-25 19:05:21 +02:00

.gitattributes

initial commit

2025-12-14 19:26:27 +00:00

.gitignore

edit_api.py:

2026-06-22 02:04:28 +02:00

AGENTS.md

dphn/Dolphin3.0-Mistral-24B is the ungated mirror of the Dolphin 3.0 Mistral 24B — exactly what you asked for. It's ~48GB fp16, which needs GPU+CPU split (device_map="auto" with 32GB on GPU, ~16GB in RAM). Let me kick off the download and update the service in parallel.

2026-06-24 21:27:11 +02:00

architecture2.svg

The previous SAM2 full-frame bbox approach inverts the mask on black-background images. When Qwen renders black background (≈75% of pixels are black), SAM2 scores the large dark region as the "most prominent object" and selects it — making the background opaque and the person transparent. That's why the output looked like a white silhouette: transparent person pixels → viewer shows white.

2026-06-22 22:09:18 +02:00

architecturev2.svg

aa

2026-06-21 04:21:32 +02:00

backlog.md

aa

2026-06-25 19:05:21 +02:00

pose.md

dphn/Dolphin3.0-Mistral-24B is the ungated mirror of the Dolphin 3.0 Mistral 24B — exactly what you asked for. It's ~48GB fp16, which needs GPU+CPU split (device_map="auto" with 32GB on GPU, ~16GB in RAM). Let me kick off the download and update the service in parallel.

2026-06-24 03:30:33 +02:00

requirements.txt

aa

2026-06-18 23:26:30 +02:00

Description

No description provided

177 MiB

Languages

HTML 62.5%

Python 35.3%

Shell 2.2%