NemoStation
/

Marlin-2B-MLX-8bit

Video-Text-to-Text

video-captioning

temporal-grounding

8-bit precision

8-bit precision

Model card Files Files and versions

Marlin-2B-MLX-8bit

2.68 GB

Ctrl+K

Ctrl+K

1 contributor

History: 4 commits

HappyPablo's picture

fix(processor): Qwen3VL*->Qwen2VL* image/video processor classes (match base + preprocessor_config; fixes AutoProcessor on sglang-mlx)

1434587 verified 13 days ago

.gitattributes

1.57 kB
Marlin-2B MLX 8-bit weights (mlx-vlm convert, --q-bits 8) 23 days ago
README.md

1.88 kB
model card 23 days ago
chat_template.jinja

7.76 kB
Marlin-2B MLX 8-bit weights (mlx-vlm convert, --q-bits 8) 23 days ago
config.json

3.73 kB
Marlin-2B MLX 8-bit weights (mlx-vlm convert, --q-bits 8) 23 days ago
generation_config.json

137 Bytes
Marlin-2B MLX 8-bit weights (mlx-vlm convert, --q-bits 8) 23 days ago
model.safetensors

2.66 GB
xet

Marlin-2B MLX 8-bit weights (mlx-vlm convert, --q-bits 8) 23 days ago
model.safetensors.index.json

81.7 kB
Marlin-2B MLX 8-bit weights (mlx-vlm convert, --q-bits 8) 23 days ago
modeling_marlin.py

23.1 kB
Marlin-2B MLX 8-bit weights (mlx-vlm convert, --q-bits 8) 23 days ago
preprocessor_config.json

390 Bytes
Marlin-2B MLX 8-bit weights (mlx-vlm convert, --q-bits 8) 23 days ago
processor_config.json

991 Bytes
fix(processor): Qwen3VL*->Qwen2VL* image/video processor classes (match base + preprocessor_config; fixes AutoProcessor on sglang-mlx) 13 days ago
tokenizer.json

20 MB
xet

Marlin-2B MLX 8-bit weights (mlx-vlm convert, --q-bits 8) 23 days ago
tokenizer_config.json

1.17 kB
Marlin-2B MLX 8-bit weights (mlx-vlm convert, --q-bits 8) 23 days ago