0e63494cf3
Add fp8 support to reshape_and_cache_flash ( #6667 )
2024-07-24 18:36:52 +00:00
fea59c7712
[Bugfix][Kernel] Use int64_t for indices in fp8 quant kernels ( #6649 )
2024-07-22 14:08:30 -06:00
396d92d5e0
[Kernel][Core] Add AWQ support to the Marlin kernel ( #6612 )
2024-07-21 19:41:42 -04:00
2e26564259
[ Kernel ] FP8 Dynamic Per Token Quant - Add scale_ub ( #6593 )
...
Co-authored-by: Varun Sundar Rabindranth <varun@neuralmagic.com >
2024-07-19 18:15:26 -07:00
4cc24f01b1
[ Kernel ] Enable Dynamic Per Token fp8 ( #6547 )
2024-07-19 23:08:15 +00:00
b5241e41d9
[ Kernel ] FP8 Dynamic-Per-Token Quant Kernel ( #6511 )
...
Co-authored-by: Varun Sundar Rabindranath <varun@neuralmagic.com >
2024-07-18 01:38:35 +00:00
978aed5300
[Kernel][Attention] Separate Attention.kv_scale into k_scale and v_scale ( #6081 )
2024-07-16 15:31:32 -07:00
b675069d74
[ Misc ] Refactor Marlin Python Utilities ( #6082 )
...
Co-authored-by: Robert Shaw <rshaw@neuralmagic.com >
2024-07-11 15:40:11 +00:00
543aa48573
[Kernel] Correctly invoke prefill & decode kernels for cross-attention (towards eventual encoder/decoder model support) ( #4888 )
...
Co-authored-by: Woosuk Kwon <woosuk.kwon@berkeley.edu >
2024-07-08 17:12:15 +00:00
69ec3ca14c
[Kernel][Model] logits_soft_cap for Gemma2 with flashinfer ( #6051 )
...
Co-authored-by: Simon Mo <simon.mo@hey.com >
2024-07-04 16:35:51 -07:00
47f0954af0
[Kernel] Expand FP8 support to Ampere GPUs using FP8 Marlin ( #5975 )
2024-07-03 17:38:00 +00:00
482045ee77
[hardware][misc] introduce platform abstraction ( #6080 )
2024-07-02 20:12:22 -07:00
7c008c51a9
[ Misc ] Refactor MoE to isolate Fp8 From Mixtral ( #5970 )
...
Co-authored-by: Robert Shaw <rshaw@neuralmagic>
Co-authored-by: Michael Goin <michael@neuralmagic.com >
2024-07-02 21:54:35 +00:00
12a59959ed
[Bugfix] adding chunking mechanism to fused_moe to handle large inputs ( #6029 )
2024-07-01 21:08:29 +00:00
614aa51203
[misc][cuda] use nvml to avoid accidentally cuda initialization ( #6007 )
2024-06-30 20:07:34 -07:00
6a2d659d28
[Bugfix] Fix compute datatype for cutlass 3.x epilogues ( #5931 )
2024-06-28 17:10:34 +00:00
57f09a419c
[Hardware][Intel] OpenVINO vLLM backend ( #5379 )
2024-06-28 13:50:16 +00:00
5bfd1bbc98
[Kernel] Adding bias epilogue support for cutlass_scaled_mm ( #5560 )
...
Co-authored-by: Chih-Chieh-Yang <7364402+cyang49@users.noreply.github.com >
Co-authored-by: Lucas Wilkinson <lwilkinson@neuralmagic.com >
2024-06-26 15:16:00 +00:00
0e9164b40a
[mypy] Enable type checking for test directory ( #5017 )
2024-06-15 04:45:31 +00:00
85657b5607
[Kernel] Factor out epilogues from cutlass kernels ( #5391 )
...
Co-authored-by: Michael Goin <michael@neuralmagic.com >
Co-authored-by: youkaichao <youkaichao@gmail.com >
Co-authored-by: zifeitong <zifei.tong@parasail.io >
Co-authored-by: Robert Shaw <114415538+robertgshaw2-neuralmagic@users.noreply.github.com >
2024-06-13 11:22:19 -07:00
5467ac3196
[Kernel][Misc] Use TORCH_LIBRARY instead of PYBIND11_MODULE for custom ops ( #5047 )
2024-06-09 16:23:30 -04:00
ca3ea51bde
[Kernel] Dynamic Per-Token Activation Quantization ( #5037 )
...
Co-authored-by: Varun Sundar Rabindranath <varunsundar08@gmail.com >
Co-authored-by: Varun Sundar Rabindranath <varun@neuralmagic.com >
2024-06-07 09:36:26 -07:00
41ca62cf03
[Misc] Add CustomOp interface for device portability ( #5255 )
2024-06-05 09:18:19 -07:00
f42a006b15
[Bugfix]: During testing, use pytest monkeypatch for safely overriding the env var that indicates the vLLM backend ( #5210 )
2024-06-03 20:32:57 -07:00
cbb2f59cc8
[Kernel] Pass a device pointer into the quantize kernel for the scales ( #5159 )
2024-06-03 09:52:30 -07:00
f081c3ce4b
[Kernel] Update Cutlass fp8 configs ( #5144 )
...
Co-authored-by: Varun Sundar Rabindranath <varun@neuralmagic.com >
Co-authored-by: Robert Shaw <114415538+robertgshaw2-neuralmagic@users.noreply.github.com >
2024-06-01 08:46:07 +00:00
260d119e86
[Kernel] Refactor CUTLASS kernels to always take scales that reside on the GPU ( #5137 )
2024-06-01 06:45:32 +00:00
a22dea54d3
[Model] Support MAP-NEO model ( #5081 )
...
Co-authored-by: Zhuohan Li <zhuohan123@gmail.com >
2024-05-30 19:24:41 -07:00
8e192ff967
[Kernel][Backend][Model] Blocksparse flash attention kernel and Phi-3-Small model ( #4799 )
...
Co-authored-by: beagleski <yunanzhang@microsoft.com >
Co-authored-by: bapatra <bapatra@microsoft.com >
Co-authored-by: Barun Patra <codedecde@users.noreply.github.com >
Co-authored-by: Michael Goin <michael@neuralmagic.com >
2024-05-24 22:00:52 -07:00
a1242324c9
[Kernel] Initial Activation Quantization Support ( #4525 )
...
Co-authored-by: Varun Sundar Rabindranath <varunsundar08@gmail.com >
Co-authored-by: Varun Sundar Rabindranath <varun@neuralmagic.com >
2024-05-23 21:29:18 +00:00
6066253296
Marlin 24 prefill performance improvement (about 25% better on average) ( #4983 )
2024-05-23 02:39:27 -04:00
ee3eea0a1b
[Misc] Take user preference in attention selector ( #4960 )
2024-05-23 07:55:56 +09:00
8674f9880e
[Kernel] Fixup for CUTLASS kernels in CUDA graphs ( #4954 )
...
Pass the CUDA stream into the CUTLASS GEMMs, to avoid future issues with CUDA graphs
2024-05-22 14:10:43 +00:00
b57e6c5949
[Kernel] Add flash-attn back ( #4907 )
2024-05-19 18:11:30 -07:00
27ce85476e
[Kernel] Add marlin_24 unit tests ( #4901 )
2024-05-19 11:37:34 -04:00
33e0823de5
[Bugfix] fix rope error when load models with different dtypes ( #4835 )
2024-05-17 18:43:34 +09:00
2060e93659
[Kernel] Add w8a8 CUTLASS kernels ( #4749 )
2024-05-16 18:32:50 -04:00
5c342570d7
Add marlin unit tests and marlin benchmark script ( #4815 )
2024-05-16 09:36:49 -04:00
8a7cc254a0
Revert "[Kernel] Use flash-attn for decoding ( #3648 )" ( #4820 )
...
Lora 3 & 4 test seems to have illegal memory access failure after this commit;
[2024-05-14 23:51:18,182 E 22 22] logging.cc:101: Unhandled exception: N3c105ErrorE. what(): CUDA error: an illegal memory access was encountered
<br class="Apple-interchange-newline">
Exmaple: https://buildkite.com/vllm/ci/builds/7382#018f793d-1527-4e1c-ab59-c3a34ec55241
This reverts commit 1356df5
2024-05-15 11:52:45 +09:00
1356df53bd
[Kernel] Use flash-attn for decoding ( #3648 )
...
Co-authored-by: Woosuk Kwon <woosuk.kwon@berkeley.edu >
Co-authored-by: LiuXiaoxuanPKU <lilyliupku@gmail.com >
2024-05-13 15:50:33 -07:00
350f9e107f
[CI/Build] Move test_utils.py to tests/utils.py ( #4425 )
...
Since #4335 was merged, I've noticed that the definition of ServerRunner in the tests is the same as in the test for OpenAI API. I have moved the class to the test utilities to avoid code duplication. (Although it only has been repeated twice so far, I will add another similar test suite in #4200 which would duplicate the code a third time)
Also, I have moved the test utilities file (test_utils.py) to under the test directory (tests/utils.py), since none of its code is actually used in the main package. Note that I have added __init__.py to each test subpackage and updated the ray.init() call in the test utilities file in order to relative import tests/utils.py.
2024-05-13 23:50:09 +09:00
c833101740
[Kernel] Refactor FP8 kv-cache with NVIDIA float8_e4m3 support ( #4535 )
2024-05-09 18:04:17 -06:00
230c4b38c1
[CI/Test] fix swap test for multi gpu ( #4689 )
2024-05-08 13:14:02 -07:00
20cfcdec99
[Core][Optimization] change python dict to pytorch tensor for blocks to swap ( #4659 )
2024-05-08 12:07:05 -07:00
0f9a6e3d22
[Bugfix][Kernel] allow non-power-of-2 for prefix prefill with alibi ( #4573 )
2024-05-08 09:19:58 -07:00
63575bc2e1
[Core][Optimization] change python dict to pytorch tensor ( #4607 )
2024-05-06 21:30:27 -07:00
2a052011ca
[Kernel] Support MoE Fp8 Checkpoints for Mixtral (Static Weights with Dynamic/Static Activations) ( #4527 )
...
Follow on to #4332 to enable FP8 checkpoint loading for Mixtral and supersedes #4436 .
This PR enables the following checkpoint loading features for Mixtral:
Supports loading fp8 checkpoints for Mixtral, such as this "nm-testing/Mixtral-8x7B-Instruct-v0.1-FP8" test model
Supports static or dynamic activation quantization with static weight quantization (all per tensor)
Supports different scales for each expert weight
Supports Fp8 in QKV layer
Notes:
The Expert Gate/Router always runs at half / full precision for now.
If there are different weight scales between QKV layer (for separate QKV weights), they are re-quantized using layer.weight_scale.max() so we can have a single gemm for performance.
2024-05-04 11:45:16 -07:00
43c413ec57
[Kernel] Use flashinfer for decoding ( #4353 )
...
Co-authored-by: LiuXiaoxuanPKU <llilyliupku@gmail.com >
2024-05-03 15:51:27 -07:00
3521ba4f25
[Core][Model runner refactoring 1/N] Refactor attn metadata term ( #4518 )
2024-05-03 10:20:12 -07:00
32881f3f31
[kernel] fix sliding window in prefix prefill Triton kernel ( #4405 )
...
Co-authored-by: SangBin Cho <rkooo567@gmail.com >
2024-05-02 11:23:37 -07:00
