| .. |
|
adapter_commons
|
2f61644f6e
SPMD optimizations (#824)
|
1 month ago |
|
assets
|
653d1a08d4
feat: add support for audio models (#891)
|
3 weeks ago |
|
attention
|
032974a28a
tpu: fix TPU type api (#975)
|
2 weeks ago |
|
common
|
b3f6eeb1d2
vlm: increase the default `max_num_batched_tokens` for multimodal models (#973)
|
2 weeks ago |
|
compilation
|
0e5cf7f840
tpu: avoid dynamo guard eval overhead (#949)
|
2 weeks ago |
|
distributed
|
61103b92d4
tpu: support single and multi-host TPUs on GKE and RayServe (#970)
|
2 weeks ago |
|
endpoints
|
0f72141ca2
api: support multiple input images
|
2 weeks ago |
|
engine
|
d4e78a428b
fix: crash when cancelling a request with multi-step (#977)
|
2 weeks ago |
|
executor
|
61103b92d4
tpu: support single and multi-host TPUs on GKE and RayServe (#970)
|
2 weeks ago |
|
inputs
|
908ff753a1
fix: phi_3.5_v loading (#896)
|
3 weeks ago |
|
kv_quant
|
8a71788372
Add OLMoE (#772)
|
2 months ago |
|
lora
|
bf4a4d8516
fix: do not register punica with torch if using older torch (#948)
|
2 weeks ago |
|
modeling
|
adb6982090
models: add support for IBM Granite (PowerLM) models (#978)
|
2 weeks ago |
|
multimodal
|
2aabf8fcf7
vlm: fix errors on ragged NestedTensors (#953)
|
2 weeks ago |
|
platforms
|
9f3e7c86e2
feat: add fused Marlin MoE kernel (#934)
|
2 weeks ago |
|
plugins
|
22a4cd4595
core: fix spec decode metrics and envs circular import (#889)
|
3 weeks ago |
|
processing
|
5be6225f38
core: support multi-step scheduling w/ async post-processor (#955)
|
2 weeks ago |
|
prompt_adapter
|
2f61644f6e
SPMD optimizations (#824)
|
1 month ago |
|
quantization
|
201db10f02
models: add support for Phi3 MoE
|
2 weeks ago |
|
server
|
22a4cd4595
core: fix spec decode metrics and envs circular import (#889)
|
3 weeks ago |
|
spec_decode
|
0dfa6b60ec
core: support logprobs with multi-step scheduling (#963)
|
2 weeks ago |
|
task_handler
|
b3f6eeb1d2
vlm: increase the default `max_num_batched_tokens` for multimodal models (#973)
|
2 weeks ago |
|
transformers_utils
|
adb6982090
models: add support for IBM Granite (PowerLM) models (#978)
|
2 weeks ago |
|
triton_utils
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 months ago |
|
__init__.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 months ago |
|
_core_ext.py
|
f1ea7711bd
core: do not compile ScalarType for torch < 2.4.0 (#938)
|
2 weeks ago |
|
_custom_ops.py
|
fcfcfc65e1
quants: add triton kernels for AWQ (#946)
|
2 weeks ago |
|
_ipex_ops.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 months ago |
|
connections.py
|
c6c91edab7
ci: update & overhaul test units (#769)
|
1 month ago |
|
constants.py
|
2f61644f6e
SPMD optimizations (#824)
|
1 month ago |
|
py.typed
|
1c988a48b2
fix logging and add py.typed
|
1 year ago |
|
scalar_type.py
|
f1d0b77c92
[0.6.0] Release Candidate (#481)
|
4 months ago |
|
version.py
|
cbd51a208a
ci: bump to 0.6.5 (#964)
|
2 weeks ago |
AlpinDale
adb6982090
models: add support for IBM Granite (PowerLM) models (#978)