dealignai/Qwen3.5-VL-2B-8bit-MLX-CRACK at main

Qwen3.5-VL-2B-8bit-MLX-CRACK

2.69 GB

Ctrl+K

Ctrl+K

1 contributor

History: 12 commits

dealignai's picture

Add Ko-fi support section

d4bec0a verified about 1 month ago

.gitattributes

1.57 kB
CRACK 2B: FA+SSM surgery, thinking-off default, EOS fixed about 1 month ago
CRACK_SURGERY_META.json

322 Bytes
CRACK 2B: FA+SSM surgery, thinking-off default, EOS fixed about 1 month ago
README.md

6.34 kB
Add Ko-fi support section about 1 month ago
chat_template.jinja

7.76 kB
CRACK 2B: FA+SSM surgery, thinking-off default, EOS fixed about 1 month ago
config.json

2.71 kB
CRACK 2B: FA+SSM surgery, thinking-off default, EOS fixed about 1 month ago
dealign_logo.png

7.66 kB
Add dealign_logo.png about 1 month ago
dealign_mascot.png

37.2 kB
Add dealign_mascot.png about 1 month ago
generation_config.json

234 Bytes
Update: increase repetition_penalty to 1.3, add presence_penalty 1.5 per Qwen recommendation about 1 month ago
model.safetensors

2.66 GB
xet

CRACK 2B: FA+SSM surgery, thinking-off default, EOS fixed about 1 month ago
model.safetensors.index.json

77.7 kB
CRACK 2B: FA+SSM surgery, thinking-off default, EOS fixed about 1 month ago
preprocessor_config.json

390 Bytes
CRACK 2B: FA+SSM surgery, thinking-off default, EOS fixed about 1 month ago
processor_config.json

1.3 kB
CRACK 2B: FA+SSM surgery, thinking-off default, EOS fixed about 1 month ago
tokenizer.json

20 MB
xet

CRACK 2B: FA+SSM surgery, thinking-off default, EOS fixed about 1 month ago
tokenizer_config.json

9.16 kB
Fix: inject chat_template with enable_thinking support into tokenizer_config.json about 1 month ago
video_preprocessor_config.json

385 Bytes
CRACK 2B: FA+SSM surgery, thinking-off default, EOS fixed about 1 month ago
vmlx-banner.png

4.15 kB
Add vMLX promotional banner to README about 1 month ago
vmlx-icon.png

1.26 kB
Add vMLX promotional banner to README about 1 month ago
vocab.json

6.72 MB
CRACK 2B: FA+SSM surgery, thinking-off default, EOS fixed about 1 month ago