ryang a6ae3af15e Support XiaomiMiMo inference with mtp (#6059)

2025-05-22 14:14:49 -07:00

..

configs

Support XiaomiMiMo inference with mtp (#6059)

2025-05-22 14:14:49 -07:00

connector

feat(remote_model): support variable remote backend for model loader (#3964)

2025-03-14 00:40:44 -07:00

constrained

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

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

disaggregation

Fix missing http status import for PD failure handler (#6520)

2025-05-22 15:23:54 +08:00

distributed

Use monotonic clock for interval measurement (#6211)

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

entrypoints

chore: upgrade sgl-kernel v0.1.4 (#6532)

2025-05-22 13:28:16 -07:00

layers

[RL] Remove the w13 weight_scale and input_scale for UnquantizedEPMoE… (#6308)

2025-05-21 22:03:15 -07:00

lora

Fix lora bench (#6302)

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

managers

Support dynamically rebalancing experts using EPLB (#6469)

2025-05-21 23:13:21 -07:00

mem_cache

[Fix]Fix capture fail bug for DeepSeek (#6275)

2025-05-21 11:11:20 -07:00

metrics

Use monotonic clock for interval measurement (#6211)

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

model_executor

Support XiaomiMiMo inference with mtp (#6059)

2025-05-22 14:14:49 -07:00

model_loader

Reland tiny refactor DefaultModelLoader.Source (#6041)

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

models

Support XiaomiMiMo inference with mtp (#6059)

2025-05-22 14:14:49 -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

Apply constraint grammar to EAGLE (#6499)

2025-05-21 17:18:41 -07:00

_custom_ops.py

Revert "fix some typos" (#6244)

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

aio_rwlock.py

A better aio rwlock that guarantees the order (#2547)

2024-12-22 15:44:32 -08:00

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

Clean up imports (#5467)

2025-04-16 15:26:49 -07:00

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

Make torch TP composable with torchao (#2436)

2024-12-11 04:21:42 -08:00

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

Fix torch.compile cacheing (#5259)

2025-04-10 18:08:45 -07:00

reasoning_parser.py

Add typo checker in pre-commit (#6179)

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

server_args.py

Support dynamically rebalancing experts using EPLB (#6469)

2025-05-21 23:13:21 -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

Support penalty in overlap mode; return logprob with chunked prefill; improve benchmark scripts (#3988)

2025-03-03 00:12:04 -08:00