| .. |
|
aqlm
|
3bdeb3e116
fix: clang formatting for all kernels (#558)
|
7 mesi fa |
|
autoquant
|
0307da9e15
refactor: bitsandbytes -> autoquant
|
7 mesi fa |
|
awq
|
9d81716bfd
[v0.5.3] Release Candidate (#388)
|
10 mesi fa |
|
compressed_tensors
|
f4ea11b982
feat: initial support for activation quantization
|
7 mesi fa |
|
cutlass_w8a8
|
f2c6791527
feat: update cutlass fp8 configs
|
7 mesi fa |
|
exl2
|
9d81716bfd
[v0.5.3] Release Candidate (#388)
|
10 mesi fa |
|
fp8
|
3bdeb3e116
fix: clang formatting for all kernels (#558)
|
7 mesi fa |
|
gguf
|
9d81716bfd
[v0.5.3] Release Candidate (#388)
|
10 mesi fa |
|
gptq
|
3bdeb3e116
fix: clang formatting for all kernels (#558)
|
7 mesi fa |
|
gptq_marlin
|
3bdeb3e116
fix: clang formatting for all kernels (#558)
|
7 mesi fa |
|
int8_kvcache
|
9810daa699
feat: INT8 KV Cache (#298)
|
1 anno fa |
|
marlin
|
d8667fcb98
improve gptq_marlin_24 prefill performance
|
7 mesi fa |
|
quip
|
aebd68c632
feat: backport kernels (#235)
|
1 anno fa |
|
squeezellm
|
8fa608aeb7
feat: replace Ray with NCCL for control plane comms (#221)
|
1 anno fa |
|
quant_ops.cpp
|
f4ea11b982
feat: initial support for activation quantization
|
7 mesi fa |
|
quant_ops.h
|
f4ea11b982
feat: initial support for activation quantization
|
7 mesi fa |
AlpinDale
0307da9e15
refactor: bitsandbytes -> autoquant