 AlpinDale
|
e75daadfbd
don't copy LogitsProcessors
|
il y a 10 mois |
|
AlpinDale
|
fd2c06e6b4
defensively copy
|
il y a 10 mois |
|
AlpinDale
|
5638e08d83
Merge branch 'main' into sampler_order_v2
|
il y a 10 mois |
|
AlpinDale
|
16615784b3
fix: prefix cache for turing gpus
|
il y a 10 mois |
 AlpinDale
|
7dc73a779a
fix: properly perform garbage collection for lora (#277)
|
il y a 10 mois |
|
AlpinDale
|
697c06c4f5
fix: LoRA support for mixtral (#276)
|
il y a 10 mois |
|
AlpinDale
|
4b80b42362
fix: memory leaks due to nccl cuda graphs (#275)
|
il y a 10 mois |
|
AlpinDale
|
e31c6f0b45
feat: refactor modeling logic and support more models (#274)
|
il y a 10 mois |
|
AlpinDale
|
7d6ba53602
feat: fused top-k kernels for MoE (#273)
|
il y a 10 mois |
|
AlpinDale
|
a3cab09b69
chore: logging env variable
|
il y a 10 mois |
|
AlpinDale
|
2c08aa5af4
chore: remove eos token from output (#272)
|
il y a 10 mois |
|
AlpinDale
|
8e1cd54497
fix: do not include fp8 for rocm (#271)
|
il y a 10 mois |
|
AlpinDale
|
6a63ab4ec3
fix: remote encode request if using ray (#270)
|
il y a 10 mois |
|
AlpinDale
|
224b87b484
feat: add fused mixtral moe support (#238)
|
il y a 10 mois |
 Thomas Xin
|
43cf0e98a0
fix: worker initialization on WSL (#260)
|
il y a 10 mois |
 swadical
|
0527131e93
fix: grammar logits processor (#268)
|
il y a 10 mois |
 Stefan Gligorijevic
|
a41366fea1
multiple fixes
|
il y a 10 mois |
|
AlpinDale
|
2370dbcfd8
feat: OPT model support (#266)
|
il y a 10 mois |
|
AlpinDale
|
6b8faa7ee8
yapf
|
il y a 10 mois |
|
AlpinDale
|
177646864a
Merge branch 'main' into sampler_order_v2
|
il y a 10 mois |
|
AlpinDale
|
4360684667
fix: cuda version in wheel
|
il y a 10 mois |
 TearGosling
|
80e8a14949
feat: add pygchat Jinja template (#218)
|
il y a 11 mois |
|
Stefan Gligorijevic
|
af72d6bf73
silly oversight
|
il y a 11 mois |
|
Stefan Gligorijevic
|
a744bf95d9
better(and correct) fix for quad
|
il y a 11 mois |
|
Stefan Gligorijevic
|
bc040c78b7
only apply quadratic sampling when requested
|
il y a 11 mois |
|
Stefan Gligorijevic
|
0cfc7e364f
properly set default order
|
il y a 11 mois |
|
Stefan Gligorijevic
|
57d822b4c8
another bit missed in merge conflict resolution
|
il y a 11 mois |
|
Stefan Gligorijevic
|
d6282a2f65
incorrectly resolved merge
|
il y a 11 mois |
|
Stefan Gligorijevic
|
916b252524
fix more missing bits
|
il y a 11 mois |
|
Stefan Gligorijevic
|
53017ce4f3
fix? sampling params for quadratic
|
il y a 11 mois |
