cao1zhg b6dd4bcb81 feat: update support for qwen3next model (#10466)

2025-09-16 16:09:56 +08:00

..

attention

feat: update support for qwen3next model (#10466)

2025-09-16 16:09:56 +08:00

moe

Fix trtllm_moe wrong correction bias (#10440)

2025-09-15 01:02:05 -07:00

quantization

Cache the result of is_blackwell platform check (#10498)

2025-09-15 22:30:28 -07:00

activation.py

model: support Apertus (#9774)

2025-09-11 20:49:10 -07:00

amx_utils.py

[CPU] refine CPU integration code (#7647)

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

communicator.py

Optimized deepseek-v3/r1 model performance on mxfp4 run (#10008)

2025-09-04 15:11:22 -07:00

dp_attention.py

Temporay work-around for rocm 7.0.0 alpha with enabling data-parallel issue (#10434)

2025-09-15 19:08:04 -07:00

elementwise.py

Update grok.py and tiktoken tokenizer (#9532)

2025-08-23 05:40:18 -07:00

flashinfer_comm_fusion.py

Support trtllm_allreduce_fusion in flashinfer for cuda<12.8 (#9339)

2025-08-20 16:54:30 -07:00

layernorm.py

Refactor fused_add_rmsnorm import logic (#10207)

2025-09-09 00:23:58 -07:00

linear.py

Support LingV2 model (#10359)

2025-09-11 23:53:52 -07:00

logits_processor.py

[Generative Score API] Scoring(Prefill-only) optimizations. (#9748)

2025-09-14 01:57:06 +08:00

model_parallel.py

Move parsers under a single folder (#9912)

2025-09-02 18:25:04 -07:00

multimodal.py

[VLM] Improving multimodal tensor hash kernel (#9008)

2025-08-14 13:45:55 -07:00

parameter.py

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

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

pooler.py

Feat/support rerank (#6058)

2025-06-16 10:50:01 -07:00

radix_attention.py

Update grok.py and tiktoken tokenizer (#9532)

2025-08-23 05:40:18 -07:00

rocm_linear_utils.py

Optimized deepseek-v3/r1 model performance on mxfp4 run (#10008)

2025-09-04 15:11:22 -07:00

rotary_embedding.py

Remove mrope position sync (#9460)

2025-09-03 11:40:53 -07:00

sampler.py

[Generative Score API] Scoring(Prefill-only) optimizations. (#9748)

2025-09-14 01:57:06 +08:00

torchao_utils.py

Support loading of larger models with on-the-fly quantization (#3061)

2025-01-22 21:33:17 -08:00

utils.py

move is_sm90_supported/is_sm100_supported to python/sglang/srt/utils.py (#9679)

2025-08-27 03:34:29 -07:00

vocab_parallel_embedding.py

Revert "Revert the changes on NCCL symmetric memory" (#10238)

2025-09-09 12:11:49 -07:00