Compat with latest VLLM 0.4.2 main + fork.number rename + Flashinfer 0.0.4 - #380

Merged

merrymercy merged 20 commits into

sgl-project:mainfrom

Qubitium:fork-cleanup

May 11, 2024

Commits on Apr 22, 2024

rename fork.number param to "size" for clarity and add add assert check
Qubitium
committed
25ed150
move parallel_utils into vllm/distributed
ZX
committed
e9261d5
import sglang.srt.weight_utils instead of vllm.model_executor.weight_utils
ZX
committed
77b3716
compat to vllm master
ZX
committed
675203c
added weight_utils.py
ZX
committed
2966f4b

Commits on Apr 23, 2024

Commits on Apr 24, 2024

Merge branch 'main' into fork-cleanup
ZX-ModelCloud
authored
02356ce

Commits on May 6, 2024

Adapt vllm PR changes: [Misc][Refactor] Generalize linear_method to be quant_method
ZX
committed
84058cb
fix vllm.logger import error
ZX
committed
aa6e47f
add comment
ZX
committed
fa80304
Removed temporary code
ZX
committed
42edf6e

Commits on May 7, 2024

Merge branch 'fork-cleanup' of https://github.com/Qubitium/sglang into fork-cleanup
Qubitium
committed
272d790
flashinfer 0.0.4 check
Qubitium
committed
06710e9
show min version in pkg exception
Qubitium
committed
95a64a7
Execute assert_pkg_version("flashinfer", "0.0.4") only once
ZX
committed
0a09c65
Merge remote-tracking branch 'origin/fork-cleanup' into fork-cleanup
ZX
committed
ed2ef49
cleanup
ZX
committed
8d79609

Commits on May 11, 2024

Merge branch 'main' into fork-cleanup
merrymercy
authored
d19d35e