 unicornchan
|
e968fa8d72
|
[feature] add flash_attn to requirements
|
2025-02-10 01:52:39 +00:00 |
|
|
liam
|
fd481af193
|
⚡ update v0.3 preview
|
2025-02-10 09:49:56 +08:00 |
|
|
liam
|
6dd4fa0e87
|
⚡ improve readme
|
2025-02-10 09:38:26 +08:00 |
|
|
unicornchan
|
fd8037cda1
|
Merge branch 'feat-DeepSeekV3' of github.com:kvcache-ai/ktrans_v0.2_dev into feat-DeepSeekV3
|
2025-02-10 01:01:14 +00:00 |
|
|
unicornchan
|
c7e6d09068
|
[feature] update version and github action jobs for package
|
2025-02-10 01:00:57 +00:00 |
|
|
chenht2022
|
2d684ee96a
|
Small fix
|
2025-02-09 16:25:43 +00:00 |
|
|
chenht2022
|
6b33f41de4
|
Add V0.3-preview doc
|
2025-02-09 16:08:16 +00:00 |
|
|
liam
|
098602b08f
|
⚡ v0.2 ongoing
|
2025-02-09 22:41:14 +08:00 |
|
|
liam
|
bf1d413be0
|
Merge branch 'feat-DeepSeekV3' of github.com:kvcache-ai/ktransformers into feat-DeepSeekV3
|
2025-02-08 13:17:10 +08:00 |
|
|
liam
|
c18ecd7b7f
|
⚡ add flush print in local_chat output and change default optimize yaml of deepseekv3 to single gpu
|
2025-02-08 13:15:52 +08:00 |
|
|
Azure
|
c4d9bc6670
|
support KExpertsMarlin backend
|
2025-02-07 05:57:40 +00:00 |
|
|
liam
|
0262f954c7
|
Merge branch 'feat-DeepSeekV3' of github.com:kvcache-ai/ktransformers into feat-DeepSeekV3
|
2025-02-06 22:41:25 +08:00 |
|
|
liam
|
3dca28d23b
|
⚡ fix moe.cpp int overflow problem
|
2025-02-06 22:39:16 +08:00 |
|
|
Azure
|
027b11266c
|
modify moeinfer param
|
2025-02-06 14:07:38 +00:00 |
|
|
Azure
|
ee24a27001
|
update v3 single gpu rule yaml;
|
2025-02-04 16:14:35 +00:00 |
|
|
Azure
|
907251c743
|
done support deepseekv3
|
2025-02-04 15:53:38 +00:00 |
|
|
Azure
|
f748cd29f0
|
fix rope; update moegate
|
2025-02-01 18:05:45 +00:00 |
|
|
Azure
|
f873558a89
|
update rope calculation; update modeling.py; update gate for moe
|
2025-02-01 07:32:21 +00:00 |
|
|
Azure
|
5a50b34627
|
fix hard coding caused by rope dim calculation, load from config now
|
2025-01-31 15:25:50 +00:00 |
|
|
Azure
|
476b1d8dc6
|
support deepseekv3; runable but have precition problem
|
2025-01-31 08:27:24 +00:00 |
|
|
UnicornChan
|
de7e892f72
|
Merge pull request #115 from KMSorSMS/main
⚡ rm opt config path default value and fix some config logic bug
|
2024-11-14 20:07:12 +08:00 |
|
|
liam
|
04cebec4bb
|
⚡ rm opt config path default value and fix some config logic bug
|
2024-11-14 20:02:30 +08:00 |
|
|
UnicornChan
|
dddc42038d
|
Merge pull request #110 from KMSorSMS/main
refactor local_chat & config setting
|
2024-11-06 09:44:12 +08:00 |
|
|
liam
|
28ae7e24ed
|
✨: finish patch
|
2024-11-04 14:05:03 +08:00 |
|
|
liam
|
c2b4dc805c
|
🚑️:roll back transformer.py and find that it's multiple chat hsitory have minor accurate error
|
2024-11-04 14:02:19 +08:00 |
|
|
liam
|
a148da2cfe
|
✨: rm sensitive info in config.yaml, add readme of makefile. support old model_path config
|
2024-11-04 14:02:19 +08:00 |
|
|
anyanqilin
|
9a2e7057c8
|
wjh fix change
|
2024-11-04 14:02:19 +08:00 |
|
|
anyanqilin
|
a72dc6ed15
|
wjh change
|
2024-11-04 14:02:19 +08:00 |
|
|
anyanqilin
|
2d67016d14
|
wjh-change
|
2024-11-04 14:02:19 +08:00 |
|
|
liam
|
7c94df4bcf
|
🚑️: back transformer.py bugs version, and fix typo error in local_chat.py
|
2024-11-04 14:02:19 +08:00 |
|
|
liam
|
dd1d8667f3
|
✨: refactor local_chat and fix message slice bug in server
|
2024-11-04 14:02:19 +08:00 |
|
|
UnicornChan
|
a6a1cc054f
|
Merge pull request #106 from Azure-Tang/main
[Fix] Fix readme structure.
|
2024-10-30 16:36:59 +08:00 |
|
|
TangJingqi
|
c7d62a67db
|
update supported models
|
2024-10-28 14:37:10 +08:00 |
|
|
TangJingqi
|
d8ddaf0ea0
|
fix readme
|
2024-10-28 14:21:12 +08:00 |
|
|
Chen Hongtao
|
43fc7f44a6
|
Merge pull request #99 from chenht2022/main
Adapt Windows
|
2024-10-09 19:09:58 +08:00 |
|
|
chenht2022
|
14869b55ad
|
Adapt Windows
|
2024-10-09 11:08:32 +00:00 |
|
|
UnicornChan
|
a81a7ffe21
|
Merge pull request #77 from TKONIY/fix-prefill-and-generate
Fix: Wrong type of token list returned by prefill_and_generate
|
2024-10-09 19:04:27 +08:00 |
|
|
Chen Hongtao
|
b4904537e3
|
Merge pull request #83 from sayap/task-queue-cond-var
Use cond var to avoid busy loop
|
2024-10-09 18:57:17 +08:00 |
|
|
UnicornChan
|
43e8848ddd
|
Merge pull request #86 from xhedit/main
typo fix: KMisrtal -> KMistral
|
2024-09-19 10:37:55 +08:00 |
|
|
UnicornChan
|
49539ac424
|
Merge pull request #88 from Azure-Tang/main
[fix] Fix some gpu dequant function doesn't support multi gpu bug
|
2024-09-15 15:17:05 +08:00 |
|
|
Azure
|
7953bd9c0d
|
update readme
|
2024-09-13 08:47:35 +00:00 |
|
|
Azure
|
3758afb526
|
fix some dequant function dosen't support multi gpu bug
|
2024-09-13 08:34:23 +00:00 |
|
|
xhedit
|
234faf7987
|
typo fix: KMisrtal -> KMistral
|
2024-09-12 15:58:01 +00:00 |
|
|
Yap Sok Ann
|
6666d62237
|
Use cond var to avoid busy loop
|
2024-09-11 16:10:54 +07:00 |
|
|
UnicornChan
|
3ed8a0437f
|
Merge pull request #72 from sayap/dequantize-iq4-xs
Support IQ4_XS dequantize
|
2024-09-06 10:06:10 +08:00 |
|
|
yangshen
|
ee72cee050
|
Fix: the tokens return by prefill_and_generate
|
2024-09-05 05:29:23 +00:00 |
|
|
UnicornChan
|
be81269e38
|
Merge pull request #71 from Azure-Tang/main
[fix] Fix qlen > chunk_size mask is none error
|
2024-09-02 14:51:05 +08:00 |
|
|
Azure
|
c55de02f7b
|
fix qlen > 1000 mask is none error
|
2024-09-02 02:58:10 +00:00 |
|
|
Yap Sok Ann
|
be356c1b8d
|
Support IQ4_XS dequantize
|
2024-09-02 09:10:19 +07:00 |
|
|
UnicornChan
|
022b893819
|
Merge pull request #67 from UnicornChan/main
[fix] fix bugs about Qwen2-57B, install requirement, DockerFile
v0.1.4
|
2024-08-30 17:57:49 +08:00 |
|