Sayankotor
/

qwen35-27b-FastKron-2bit

Model card Files Files and versions

qwen35-27b-FastKron-2bit

12.2 GB

Ctrl+K

Ctrl+K

1 contributor

History: 15 commits

Sayankotor's picture

codebook_init: guard torch.library.define against re-registration (multi-ckpt in one kernel)

b7417d8 verified 12 days ago

.gitattributes

1.57 kB
compressed 2-bit QTIP (400/400 proj, vision fp16) 13 days ago
bitshift.py

22.6 kB
bitshift: importlib lazy imports (check_imports-free, on-the-fly path needs no kernels) 12 days ago
chat_template.jinja

7.76 kB
compressed 2-bit QTIP (400/400 proj, vision fp16) 13 days ago
codebook_init.py

4.18 kB
codebook_init: guard torch.library.define against re-registration (multi-ckpt in one kernel) 12 days ago
config.json

7.8 kB
Add auto_map for trust_remote_code 12 days ago
generation_config.json

213 Bytes
compressed 2-bit QTIP (400/400 proj, vision fp16) 13 days ago
kernel_check.py

845 Bytes
trust_remote_code: kernel_check.py 12 days ago
kernel_decompress.py

2.24 kB
trust_remote_code: kernel_decompress.py 12 days ago
matmul_had.py

1.66 MB
trust_remote_code: matmul_had.py 12 days ago
misc.py

610 Bytes
trust_remote_code: misc.py 12 days ago
model.safetensors

12.2 GB
xet

compressed 2-bit QTIP (400/400 proj, vision fp16) 13 days ago
modeling_qwen3_5_quantized.py

6.32 kB
trust_remote_code: modeling_qwen3_5_quantized.py 12 days ago
quantized_linear.py

5.32 kB
trust_remote_code: quantized_linear.py 12 days ago
quip_params.pt

4.55 kB
xet

compressed 2-bit QTIP (400/400 proj, vision fp16) 13 days ago
tokenizer.json

20 MB
xet

compressed 2-bit QTIP (400/400 proj, vision fp16) 13 days ago
tokenizer_config.json

1.1 kB
compressed 2-bit QTIP (400/400 proj, vision fp16) 13 days ago