jianan-gu d389bedf72 [CPU][Qwen3 MoE] Enable fused_topk CPU fusion and enhance FP8 TP padding (#7838)

2025-07-09 02:04:21 -07:00

..

configs

feat: support DeepSeek-R1-W4AFP8 model with ep-moe mode (#7762)

2025-07-07 14:47:21 -07:00

connector

…

constrained

…

disaggregation

[PD] Add guidance for prefill bootstrap timeout (#7846)

2025-07-08 21:00:34 -07:00

distributed

feat: use D2D instead of H2H in pp (#7673)

2025-07-03 10:58:50 -07:00

entrypoints

[Multimodal][Perf] Use pybase64 instead of base64 (#7724)

2025-07-08 14:00:58 -07:00

eplb

Move files related to EPLB (#7580)

2025-06-29 15:39:38 -07:00

function_call

feat(func_call): Add more check in BaseFormatDetector.parse_streaming_increment (#7479)

2025-06-23 11:15:47 -07:00

layers

[CPU][Qwen3 MoE] Enable fused_topk CPU fusion and enhance FP8 TP padding (#7838)

2025-07-09 02:04:21 -07:00

lora

Fix division-by-zero bug in LoRA triton kernels. (#7785)

2025-07-06 00:45:29 -07:00

managers

fix: Handles input_embeds in GenerateReqInput when n>1 (#7830)

2025-07-08 14:00:42 -07:00

mem_cache

Kernels for efficient KV cache IO (#7313)

2025-07-06 22:53:36 -07:00

metrics

…

model_executor

DP Attention with Auto DeepEP Dispatch (#7222)

2025-07-05 01:54:24 -07:00

model_loader

[CPU] support the case where num_attention_heads or intermediate_size is not divisible by the TP size (#6771)

2025-07-03 09:51:38 -07:00

models

Revert "Embedding parallel by attn_tp (#7623)" (#7880)

2025-07-08 22:03:09 -07:00

multimodal

[Multimodal][Perf] Use pybase64 instead of base64 (#7724)

2025-07-08 14:00:58 -07:00

sampling

[Minor] clean up multimodal processor and tokenizer manager (#7624)

2025-06-29 02:50:14 -07:00

speculative

Add treemask mode to build_eagle_tree & release sgl-kernel 0.2.3 (#7756)

2025-07-05 12:17:05 -07:00

_custom_ops.py

…

aio_rwlock.py

…

code_completion_parser.py

feat(oai refactor): Replace openai_api with entrypoints/openai (#7351)

2025-06-21 13:21:06 -07:00

constants.py

…

conversation.py

Support Mimo-VL (#7579)

2025-07-08 14:00:25 -07:00

custom_op.py

npu fused op (#7386)

2025-06-25 01:54:20 -07:00

debug_utils.py

…

hf_transformers_utils.py

Move mem_fraction_static adjustment for multimodal models to server_args.py & Fix session control & Other cleanups (#7748)

2025-07-04 16:33:33 -07:00

jinja_template_utils.py

feat(oai refactor): Replace openai_api with entrypoints/openai (#7351)

2025-06-21 13:21:06 -07:00

model_parallel.py

…

operations_strategy.py

feat: support compatibility between MTP and two-batch-overlap (#7225)

2025-06-27 01:10:27 -07:00

operations.py

…

patch_torch.py

…

reasoning_parser.py

Fix stream reasoning parser and Adds Kimi reasoning parser (#7432)

2025-06-29 14:39:05 -07:00

server_args.py

Fixes typo in assertion message (#7895)

2025-07-09 01:47:14 -07:00

torch_memory_saver_adapter.py

…

two_batch_overlap.py

[Bugfix] Fix two batch overlap with auto DeepEP Dispatch (#7853)

2025-07-08 02:49:32 -07:00

utils.py

[Multimodal][Perf] Use pybase64 instead of base64 (#7724)

2025-07-08 14:00:58 -07:00

warmup.py

Improve streaming, log_level, memory report, weight loading, and benchmark script (#7632)

2025-06-29 23:16:19 -07:00