| .. |
|
compressed_tensors
|
f5bbf07c90
chore: use the `compressed-tensors` library to avoid code reuse (#704)
|
пре 6 месеци |
|
gguf_utils
|
9d81716bfd
[v0.5.3] Release Candidate (#388)
|
пре 10 месеци |
|
utils
|
3170c0d4c6
fix: GPTQ/AWQ on Colab (#655)
|
пре 6 месеци |
|
__init__.py
|
3f49a55f82
feat: add INT8 W8A16 quant for TPU (#663)
|
пре 6 месеци |
|
aqlm.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
пре 6 месеци |
|
autoquant.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
пре 6 месеци |
|
awq.py
|
edec2e9a9e
feat: migrate awq and awq_marlin to AphroditeParameter (#702)
|
пре 6 месеци |
|
awq_marlin.py
|
edec2e9a9e
feat: migrate awq and awq_marlin to AphroditeParameter (#702)
|
пре 6 месеци |
|
base_config.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
пре 6 месеци |
|
bitsandbytes.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
пре 6 месеци |
|
deepspeedfp.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
пре 6 месеци |
|
eetq.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
пре 6 месеци |
|
exl2.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
пре 6 месеци |
|
fbgemm_fp8.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
пре 6 месеци |
|
fp8.py
|
4ec08af18b
chore: update fused MoE weight loading (#700)
|
пре 6 месеци |
|
gguf.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
пре 6 месеци |
|
gptq.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
пре 6 месеци |
|
gptq_marlin.py
|
4f6020cc86
chore: migrate gptq_marlin to AphroditeParameters (#699)
|
пре 6 месеци |
|
gptq_marlin_24.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
пре 6 месеци |
|
hadamard.safetensors
|
9d81716bfd
[v0.5.3] Release Candidate (#388)
|
пре 10 месеци |
|
kv_cache.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
пре 6 месеци |
|
marlin.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
пре 6 месеци |
|
qqq.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
пре 6 месеци |
|
quip.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
пре 6 месеци |
|
quip_utils.py
|
9d81716bfd
[v0.5.3] Release Candidate (#388)
|
пре 10 месеци |
|
schema.py
|
9d81716bfd
[v0.5.3] Release Candidate (#388)
|
пре 10 месеци |
|
squeezellm.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
пре 6 месеци |
|
tpu_int8.py
|
3f49a55f82
feat: add INT8 W8A16 quant for TPU (#663)
|
пре 6 месеци |
AlpinDale
f5bbf07c90
chore: use the `compressed-tensors` library to avoid code reuse (#704)