Michael Goin 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
..
async_engine	Fix/async chat serving (#2727)	2024-05-03 11:04:14 -07:00
basic_correctness	[Kernel] Use flashinfer for decoding (#4353)	2024-05-03 15:51:27 -07:00
core	[Core] Ignore infeasible swap requests. (#4557)	2024-05-02 14:31:20 -07:00
distributed	[Kernel] Use flashinfer for decoding (#4353)	2024-05-03 15:51:27 -07:00
engine	[Core] Add multiproc_worker_utils for multiprocessing-based workers (#4357)	2024-05-01 18:41:59 +00:00
entrypoints	Fix/async chat serving (#2727)	2024-05-03 11:04:14 -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] Support MoE Fp8 Checkpoints for Mixtral (Static Weights with Dynamic/Static Activations) (#4527)	2024-05-04 11:45:16 -07:00
lora	[Kernel] Full Tensor Parallelism for LoRA Layers (#3524)	2024-04-27 00:03:48 -07:00
metrics	[CI]Add regression tests to ensure the async engine generates metrics (#4524)	2024-05-01 19:57:12 -07:00
model_executor	[Core] Support offline use of local cache for models (#4374)	2024-04-27 09:59:55 -07:00
models	[Kernel] Support running GPTQ 8-bit models in Marlin (#4533)	2024-05-02 12:56:22 -04:00
prefix_caching	[Core][Bugfix]Refactor block manager for better testability (#3492)	2024-03-27 23:59:28 -07:00
prompts	[BugFix] Fix input positions for long context with sliding window (#2088)	2023-12-13 12:28:13 -08:00
quantization	[Kernel] Marlin Expansion: Support AutoGPTQ Models with Marlin (#3922)	2024-04-29 09:35:34 -07:00
samplers	[Core][Model runner refactoring 1/N] Refactor attn metadata term (#4518)	2024-05-03 10:20:12 -07:00
spec_decode	[Misc][Refactor] Introduce ExecuteModelData (#4540)	2024-05-03 17:47:07 -07:00
tensorizer_loader	[Core][Distributed] use cpu group to broadcast metadata in cpu (#4444)	2024-04-29 13:52:22 -07:00
tokenization	[Bugfix] Fix parameter name in get_tokenizer (#4107)	2024-04-25 19:10:48 -07:00
worker	[Misc][Refactor] Introduce ExecuteModelData (#4540)	2024-05-03 17:47:07 -07:00
__init__.py	[Small] Formatter only checks lints in changed files (#1528)	2023-10-31 15:39:38 -07:00
conftest.py	[Bug fix][Core] assert num_new_tokens == 1 fails when SamplingParams.n is not 1 and max_tokens is large & Add tests for preemption (#4451)	2024-05-01 19:24:13 -07:00
test_cache_block_hashing.py	[CI] Try introducing isort. (#3495)	2024-03-25 07:59:47 -07:00
test_config.py	[Core] Refactor model loading code (#4097)	2024-04-16 11:34:39 -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	[Core][Model runner refactoring 1/N] Refactor attn metadata term (#4518)	2024-05-03 10:20:12 -07:00
test_regression.py	[BugFix] Fix GC bug for LLM class (#2882)	2024-02-14 22:17:44 -08:00
test_sampling_params.py	[Bugfix] fix crash if max_tokens=None (#2570)	2024-01-23 22:38:55 -08:00
test_sequence.py	[Chunked Prefill][4/n] Chunked prefill scheduler. (#3853)	2024-04-05 10:17:58 -07:00