Cody Yu 5985e3427d [Kernel] Vectorized FP8 quantize kernel (#5396) ... Inspired by #5146, this PR improves FP8 quantize kernel by vectorizing data transfer to better utilize memory bandwidth. Microbenchmark shows that this improved kernel can achieve 1.0x-1.5x speedup (especially when hidden size is large). In details, we applied 3 optimizations: - Use inverted scale so that most divisions are changed to multiplications. - Unroll the loop by 4 times to improve ILP. - Use vectorized 4 to transfer data between HBM and SRAM.		2024-06-12 14:07:26 -07:00
..
async_engine	[Bugfix][Frontend] Cleanup "fix chat logprobs" (#5026)	2024-06-10 22:36:46 -07:00
basic_correctness	[CI/Test] improve robustness of test (vllm_runner) (#5357)	2024-06-08 08:59:20 +00:00
core	[CI] Upgrade codespell version. (#5381)	2024-06-12 10:06:14 -07:00
distributed	[Core][Distributed] add same-node detection (#5369)	2024-06-11 10:53:59 -07:00
engine	[CI/Test] improve robustness of test (vllm_runner) (#5357)	2024-06-08 08:59:20 +00:00
entrypoints	[Bugfix][Frontend] Cleanup "fix chat logprobs" (#5026)	2024-06-10 22:36:46 -07:00
fp8_kv	Enable scaled FP8 (e4m3fn) KV cache on ROCm (AMD GPU) (#3290)	2024-04-03 14:15:55 -07:00
kernels	[Kernel][Misc] Use TORCH_LIBRARY instead of PYBIND11_MODULE for custom ops (#5047)	2024-06-09 16:23:30 -04:00
lora	[Misc] Improve error message when LoRA parsing fails (#5194)	2024-06-10 19:38:49 +08:00
metrics	[CI/Test] improve robustness of test (vllm_runner) (#5357)	2024-06-08 08:59:20 +00:00
model_executor	[CI/Build] Move test_utils.py to tests/utils.py (#4425)	2024-05-13 23:50:09 +09:00
models	Revert "[CI/Build] Add is_quant_method_supported to control quantization test configurations" (#5463)	2024-06-12 10:03:24 -07:00
multimodal	[Model] Initial support for LLaVA-NeXT (#4199)	2024-06-10 12:47:15 +00:00
prefix_caching	[Bugfix / Core] Prefix Caching Guards (merged with main) (#4846)	2024-05-27 15:18:17 -07:00
prompts	[BugFix] Fix input positions for long context with sliding window (#2088)	2023-12-13 12:28:13 -08:00
quantization	[Kernel] Vectorized FP8 quantize kernel (#5396)	2024-06-12 14:07:26 -07:00
samplers	[CI/Test] improve robustness of test (vllm_runner) (#5357)	2024-06-08 08:59:20 +00:00
spec_decode	[Core][Doc] Default to multiprocessing for single-node distributed case (#5230)	2024-06-11 11:10:41 -07:00
tensorizer_loader	[Bugfix][Frontend] Cleanup "fix chat logprobs" (#5026)	2024-06-10 22:36:46 -07:00
tokenization	[Core] Support image processor (#4197)	2024-06-02 22:56:41 -07:00
worker	[Core][2/N] Model runner refactoring part 2. Combine prepare prefill / decode to a single API (#4681)	2024-05-15 14:00:10 +09:00
__init__.py	[Small] Formatter only checks lints in changed files (#1528)	2023-10-31 15:39:38 -07:00
conftest.py	[CI/Test] improve robustness of test (vllm_runner) (#5357)	2024-06-08 08:59:20 +00:00
test_cache_block_hashing.py	[Core] Avoid the need to pass None values to Sequence.inputs (#5099)	2024-05-29 16:05:01 -07:00
test_config.py	[Frontend] Customizable RoPE theta (#5197)	2024-06-11 10:42:26 -07:00
test_inputs.py	[Core] Consolidate prompt arguments to LLM engines (#4328)	2024-05-28 13:29:31 -07:00
test_logger.py	[MISC] Rework logger to enable pythonic custom logging configuration to be provided (#4273)	2024-05-01 17:34:40 -07:00
test_logits_processor.py	[Misc] Remove unnecessary ModelRunner imports (#4703)	2024-05-09 00:17:17 -07:00
test_regression.py	Bugfix: fix broken of download models from modelscope (#5233)	2024-06-06 09:28:10 -07:00
test_sampling_params.py	[Bugfix] fix crash if max_tokens=None (#2570)	2024-01-23 22:38:55 -08:00
test_sequence.py	[CI/Build] Move test_utils.py to tests/utils.py (#4425)	2024-05-13 23:50:09 +09:00
test_sharded_state_loader.py	[CI] Upgrade codespell version. (#5381)	2024-06-12 10:06:14 -07:00
test_utils.py	[Misc][Utils] allow get_open_port to be called for multiple times (#5333)	2024-06-06 22:15:11 -07:00
utils.py	[FRONTEND] OpenAI tools support named functions (#5032)	2024-06-03 18:25:29 -05:00