auto_fp8
|
add support block-wise quant from bf16
|
2025-07-01 07:59:03 +08:00 |
configs
|
Release DeepSeek-V3
|
2024-12-26 19:01:57 +08:00 |
bf16_cast_fp8.py
|
update script and verified correctness
|
2025-07-01 17:40:04 +08:00 |
convert.py
|
clarify assertion error
|
2025-01-28 13:16:54 +01:00 |
generate.py
|
clarify assertion error
|
2025-01-28 13:16:54 +01:00 |
kernel.py
|
add support block-wise quant from bf16
|
2025-07-01 07:59:03 +08:00 |
model.py
|
modify the explanation of MLA
|
2025-02-26 17:07:39 +08:00 |
requirements.txt
|
Release DeepSeek-V3
|
2024-12-26 19:01:57 +08:00 |