Pavani Majety 21176b0093 [Bugfix] Fix Weightloading for the original nvidia/Deepseek-R1-FP4 checkpoint (#9940) ...

Signed-off-by: Pavani Majety <pmajety@nvidia.com>
Co-authored-by: Yineng Zhang <me@zhyncs.com>
Co-authored-by: fzyzcjy <5236035+fzyzcjy@users.noreply.github.com>

2025-09-10 12:00:23 -07:00

..

attention

Cleaning codes for speculative attention mode (#10149)

2025-09-08 17:38:06 -07:00

moe

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

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

quantization

[Bugfix] Fix Weightloading for the original nvidia/Deepseek-R1-FP4 checkpoint (#9940)

2025-09-10 12:00:23 -07:00

activation.py

[CPU] Add gelu_and_mul kernel in sgl-kernel and add ut (#9300)

2025-09-08 23:23:13 -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

[Bugfix] fix kv buffer register & dp attention & deepepmoe (#9327)

2025-08-19 10:09:48 -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

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

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

logits_processor.py

[Bug fix] Fix Gemma 2 and fix Gemma 3 multimodal with bs > 1 on NPU (#9871)

2025-09-08 01:19:40 -07:00

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

feat: add original logprobs to response (#8375)

2025-08-29 11:43:57 -07:00

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