Azaz666
/

FastVLM-1.5B-PYTORCH-INT8

Model card Files Files and versions

FastVLM-1.5B-PYTORCH-INT8

2.3 GB

Ctrl+K

Ctrl+K

1 contributor

History: 4 commits

Azaz666's picture

Replace with per-group INT8 quantization (group_size=128, lossless vs FP16)

0140a80 2 months ago

.gitattributes

1.57 kB
Upload PyTorch INT8 quantized FastVLM-1.5B (weights + config + eval) 2 months ago
chat_template.jinja

328 Bytes
Upload PyTorch INT8 quantized FastVLM-1.5B (weights + config + eval) 2 months ago
config.json

2.08 kB
Upload PyTorch INT8 quantized FastVLM-1.5B (weights + config + eval) 2 months ago
eval_results.json

941 Bytes
Replace with per-group INT8 quantization (group_size=128, lossless vs FP16) 2 months ago
llava_qwen.py

82.3 kB
Upload PyTorch INT8 quantized FastVLM-1.5B (weights + config + eval) 2 months ago
quant_info.json

187 Bytes
Replace with per-group INT8 quantization (group_size=128, lossless vs FP16) 2 months ago
state_dict.pt
Detected Pickle imports (5)
- "torch.CharStorage",
- "torch.HalfStorage",
- "torch.LongStorage",
- "collections.OrderedDict",
- "torch._utils._rebuild_tensor_v2"
What is a pickle import?
2.29 GB
xet

Replace with per-group INT8 quantization (group_size=128, lossless vs FP16) 2 months ago
tokenizer.json

11.4 MB
xet

Upload PyTorch INT8 quantized FastVLM-1.5B (weights + config + eval) 2 months ago
tokenizer_config.json

444 Bytes
Upload PyTorch INT8 quantized FastVLM-1.5B (weights + config + eval) 2 months ago