fzyzcjy c471d39eb9 Support loading weights when physical experts are different from logical experts (#6386)

2025-05-19 21:05:53 -07:00

..

configs

refactor: minor refactors regarding multimodal processing (#6187)

2025-05-17 22:53:20 -07:00

connector

…

constrained

Support incremental streaming of logprob/token_ids between scheduler and detokenizer (#6225)

2025-05-12 14:33:38 -07:00

disaggregation

[Metrics] Add KV events publishing (#6098)

2025-05-19 14:19:54 -07:00

distributed

Use monotonic clock for interval measurement (#6211)

2025-05-17 16:49:18 -07:00

entrypoints

chore: upgrade sgl-kernel v0.1.3 (#6377)

2025-05-17 19:47:05 -07:00

layers

Support loading weights when physical experts are different from logical experts (#6386)

2025-05-19 21:05:53 -07:00

lora

Fix lora bench (#6302)

2025-05-15 10:09:55 -07:00

managers

Support loading weights when physical experts are different from logical experts (#6386)

2025-05-19 21:05:53 -07:00

mem_cache

[Metrics] Add KV events publishing (#6098)

2025-05-19 14:19:54 -07:00

metrics

Use monotonic clock for interval measurement (#6211)

2025-05-17 16:49:18 -07:00

model_executor

Expert distribution recording without overhead for EPLB (#4957)

2025-05-19 20:07:43 -07:00

model_loader

Reland tiny refactor DefaultModelLoader.Source (#6041)

2025-05-17 17:11:20 -07:00

models

Refactor DeepSeek logic into atomic operations (#6326)

2025-05-19 21:05:30 -07:00

openai_api

Revert "Implement return_hidden_states for the OpenAI API (#6137)" (#6440)

2025-05-19 18:21:29 -07:00

sampling

Skip the flaky test_stateful_custom_logit_processor (#6251)

2025-05-12 18:29:41 -07:00

speculative

Use monotonic clock for interval measurement (#6211)

2025-05-17 16:49:18 -07:00

_custom_ops.py

Revert "fix some typos" (#6244)

2025-05-12 12:53:26 -07:00

aio_rwlock.py

…

code_completion_parser.py

Add typo checker in pre-commit (#6179)

2025-05-11 12:55:00 +08:00

conversation.py

The Gemma template is missing a newline after the user role. (#6331)

2025-05-18 01:57:27 -07:00

custom_op.py

…

function_call_parser.py

Add typo checker in pre-commit (#6179)

2025-05-11 12:55:00 +08:00

hf_transformers_utils.py

refactor: minor refactors regarding multimodal processing (#6187)

2025-05-17 22:53:20 -07:00

mm_utils.py

refactor: minor refactors regarding multimodal processing (#6187)

2025-05-17 22:53:20 -07:00

model_parallel.py

…

operations_strategy.py

Refactor DeepSeek logic into atomic operations (#6326)

2025-05-19 21:05:30 -07:00

operations.py

Refactor DeepSeek logic into atomic operations (#6326)

2025-05-19 21:05:30 -07:00

patch_torch.py

…

reasoning_parser.py

Add typo checker in pre-commit (#6179)

2025-05-11 12:55:00 +08:00

server_args.py

Expert distribution recording without overhead for EPLB (#4957)

2025-05-19 20:07:43 -07:00

torch_memory_saver_adapter.py

[CI] Fix ci tests (#5769)

2025-04-27 07:18:10 -07:00

utils.py

Expert distribution recording without overhead for EPLB (#4957)

2025-05-19 20:07:43 -07:00

warmup.py

…