Commit History

UPDATE: [0] prefill
efc2316

sparkleman commited on

UPDATE
ed0f7b4

sparkleman commited on

UPDATE
6b82cc0

sparkleman commited on

TRY max_split_size_mb:128
43cae0b

sparkleman commited on

UPDATE
d3e033f

sparkleman commited on

UPDATE
7063b82

sparkleman commited on

UPDATE
1c4c774

sparkleman commited on

UPDATE: merge modelscope
b76f7cc

sparkleman commited on

UPDATE: Torch memory
d055fef

sparkleman commited on

UPDATE
3b5e872

sparkleman commited on

UPDATE: Change gpu state display
aeaf225

sparkleman commited on

UPDATE: log context
2792ede

sparkleman commited on

FIX: Remove print
a14fe00

sparkleman commited on

UPDATE: cache_word_list
4ed9fde

sparkleman commited on

UPDATE: Remove <think> tag in content & handle EOS token
94c4923

sparkleman commited on

UPDATE: change default model load workflow
50f89e3

sparkleman commited on

UPDATE: support stop tokens
9b9e15b

sparkleman commited on

FIX: typo
0de6f92

sparkleman commited on

UPDATE: Add frontend
adb6ad5

sparkleman commited on

UPDATE: Merge cuda core from BlinkDL/RWKV-Gradio-1
ff3952a

sparkleman commited on

CKPT: Space CPU version
05b6df6

sparkleman commited on

UPDATE
8af9256

sparkleman commited on

UPDATE
dac6105

sparkleman commited on

FIX: cpu fallback
271e92e

sparkleman commited on

INIT
109a0c8

sparkleman commited on