21.9 GB

Ctrl+K

1 contributor

History: 17 commits

caiovicentino1

Remove legacy polar_config.json

09aa406 verified 5 months ago

.gitattributes

1.57 kB
Config + processor 5 months ago
README.md

6.35 kB
HLWQ rebrand: title, tags, notice, self-links 5 months ago
chat_template.jinja

12 kB
Config + processor 5 months ago
compression_chart.png

56.9 kB
Upload compression_chart.png with huggingface_hub 5 months ago
config.json

4.77 kB
fix: quant_method polar -> polarengine for vLLM compatibility 5 months ago
context_chart.png

46.2 kB
Upload context_chart.png with huggingface_hub 5 months ago
hlwq_config.json

398 Bytes
Add hlwq_config.json (rename from polar_config.json) 5 months ago
kv_speed_chart.png

40.3 kB
Upload kv_speed_chart.png with huggingface_hub 5 months ago
model_vision_int4.pt
Detected Pickle imports (14)
- "torch.bfloat16",
- "collections.OrderedDict",
- "torch.BFloat16Storage",
- "torch.serialization._get_layout",
- "torch.int32",
- "torch._tensor._rebuild_from_type_v2",
- "torch.device",
- "torch.IntStorage",
- "torch._utils._rebuild_wrapper_subclass",
- "torchao.dtypes.affine_quantized_tensor.AffineQuantizedTensor",
- "torchao.quantization.quant_primitives.ZeroPointDomain",
- "torchao.dtypes.uintx.tensor_core_tiled_layout.TensorCoreTiledLayout",
- "torch._utils._rebuild_tensor_v2",
- "torchao.dtypes.uintx.tensor_core_tiled_layout.TensorCoreTiledAQTTensorImpl"
How to fix it?
21.8 GB
xet

PolarQuant Vision: INT4 text + BF16 vision 5 months ago
processor_config.json

1.69 kB
Config + processor 5 months ago
tokenizer.json

32.2 MB
xet

Config + processor 5 months ago
tokenizer_config.json

2.69 kB
Config + processor 5 months ago
vram_breakdown.png

57.4 kB
Upload vram_breakdown.png with huggingface_hub 5 months ago

Detected Pickle imports (14)