sglang 0.4.2.post2__py3-none-any.whl → 0.4.2.post4__py3-none-any.whl

sglang/srt/speculative/eagle_utils.py

@@ -4,6 +4,7 @@ import dataclasses
 from typing import TYPE_CHECKING, List
 
 import torch
+import torch.nn.functional as F
 import triton
 import triton.language as tl
 
@@ -11,7 +12,14 @@ from sglang.srt.layers.attention.flashinfer_backend import (
     create_flashinfer_kv_indices_triton,
 )
 from sglang.srt.model_executor.forward_batch_info import CaptureHiddenMode
-from sglang.srt.speculative.build_eagle_tree import build_tree_kernel
+from sglang.srt.speculative.build_eagle_tree import (
+    build_tree_kernel,
+    build_tree_kernel_efficient,
+)
+from sglang.srt.utils import is_cuda_available
+
+if is_cuda_available():
+    from sgl_kernel import tree_speculative_sampling_target_only
 
 if TYPE_CHECKING:
     from sglang.srt.managers.schedule_batch import ScheduleBatch
@@ -160,8 +168,11 @@ class EagleVerifyInput:
     custom_mask: torch.Tensor
     positions: torch.Tensor
     retrive_index: torch.Tensor
+    retrive_next_token: torch.Tensor
+    retrive_next_sibling: torch.Tensor
     retrive_cum_len: torch.Tensor
     draft_token_num: int
+    spec_steps: int
     capture_hidden_mode: CaptureHiddenMode
 
     @classmethod
@@ -175,38 +186,68 @@ class EagleVerifyInput:
         seq_lens_sum: int,
         topk: int,
         spec_steps: int,
-        num_verify_token: int,
+        num_verify_tokens: int,
+        is_all_greedy: bool,
     ):
-        score_list = torch.cat(score_list, dim=1).flatten(
-            1
-        )  # b, n, topk; n= 1 + (num_steps-1) * self.topk
-        ss_token_list = torch.cat(
-            token_list, dim=1
-        )  # b, (self.topk + (num_steps-1) * self.topk)
-        top_scores = torch.topk(score_list, num_verify_token - 1, dim=-1)
-        top_scores_index = top_scores.indices
-        top_scores_index = torch.sort(top_scores_index).values
-        draft_tokens = torch.gather(ss_token_list, index=top_scores_index, dim=1)
-        draft_tokens = torch.cat((verified_id.unsqueeze(1), draft_tokens), dim=1)
-        parent_list = torch.cat(parents_list[:-1], dim=1)
-        tree_mask, position, retrive_index, retrive_cum_len = build_tree_kernel(
-            parent_list,
-            top_scores_index,
-            seq_lens,
-            seq_lens_sum,
-            topk,
-            spec_steps,
-            num_verify_token,
-        )
-        return cls(
-            draft_tokens.flatten(),
-            tree_mask,
-            position,
-            retrive_index,
-            retrive_cum_len,
-            num_verify_token,
-            CaptureHiddenMode.FULL,
-        )
+        if is_all_greedy:
+            tree_mask, position, retrive_index, retrive_cum_len, draft_tokens = (
+                build_tree_kernel(
+                    verified_id,
+                    score_list,  # b, n, topk; n= 1 + (num_steps-1) * self.topk
+                    token_list,
+                    parents_list,
+                    seq_lens,
+                    seq_lens_sum,
+                    topk,
+                    spec_steps,
+                    num_verify_tokens,
+                )
+            )
+
+            return cls(
+                draft_tokens,
+                tree_mask,
+                position,
+                retrive_index,
+                None,
+                None,
+                retrive_cum_len,
+                num_verify_tokens,
+                spec_steps,
+                CaptureHiddenMode.FULL,
+            )
+        else:
+            (
+                tree_mask,
+                position,
+                retrive_index,
+                retrive_next_token,
+                retrive_next_sibling,
+                draft_tokens,
+            ) = build_tree_kernel_efficient(
+                verified_id,
+                score_list,
+                token_list,
+                parents_list,
+                seq_lens,
+                seq_lens_sum,
+                topk,
+                spec_steps,
+                num_verify_tokens,
+            )
+
+            return cls(
+                draft_tokens,
+                tree_mask,
+                position,
+                retrive_index,
+                retrive_next_token,
+                retrive_next_sibling,
+                None,
+                num_verify_tokens,
+                spec_steps,
+                CaptureHiddenMode.FULL,
+            )
 
     def prepare_for_verify(self, batch: ScheduleBatch):
         batch.input_ids = self.draft_token
@@ -258,39 +299,71 @@ class EagleVerifyInput:
         return kv_indices, cum_kv_seq_len, qo_indptr, self.custom_mask
 
     def verify(self, batch: ScheduleBatch, logits_output: torch.Tensor) -> torch.Tensor:
-        predict = torch.argmax(logits_output.next_token_logits, dim=-1)
-        predict = torch.cat(
-            [predict, torch.full([1], -1, dtype=torch.long, device="cuda")], dim=-1
-        )
         draft_token = torch.cat(
-            [self.draft_token, torch.full([1], -1, dtype=torch.long, device="cuda")],
+            [self.draft_token, torch.full([1], -1, dtype=torch.int32, device="cuda")],
             dim=-1,
         )
-        target_predict = predict[self.retrive_index]
         candidates = draft_token[self.retrive_index]
-        # logits = logits_output.next_token_logits[self.retrive_index]
-        # target_predict = torch.argmax(logits[:, :-1], dim=-1)
-        accept_mask = candidates[:, 1:] == target_predict[:, :-1]
-        accept_mask = (torch.cumprod(accept_mask, dim=1)).sum(dim=1)
-        bs = self.retrive_cum_len.numel() - 1
-
-        max_draft_len = self.retrive_index.shape[-1]
-        accept_index = torch.full(
-            (bs, max_draft_len), -1, dtype=torch.long, device="cuda"
-        )
-        accept_length = torch.empty((bs,), dtype=torch.int, device="cuda")
-        extract_index = torch.full((bs * 2,), 0, dtype=torch.int, device="cuda")
-        eagle_verify_retrive[(bs,)](
-            self.retrive_index.contiguous(),
-            accept_mask.contiguous(),
-            self.retrive_cum_len,
-            accept_index,
-            accept_length,
-            extract_index,
-            max_draft_len,
-            self.draft_token_num,
-            triton.next_power_of_2(max_draft_len),
-        )
+        if batch.sampling_info.is_all_greedy:
+            # temp == 0
+            bs = self.retrive_cum_len.numel() - 1
+            predict = torch.argmax(logits_output.next_token_logits, dim=-1)
+            predict = torch.cat(
+                [predict, torch.full([1], -1, dtype=torch.int32, device="cuda")], dim=-1
+            )
+            target_predict = predict[self.retrive_index]
+            # logits = logits_output.next_token_logits[self.retrive_index]
+            # target_predict = torch.argmax(logits[:, :-1], dim=-1)
+            accept_mask = candidates[:, 1:] == target_predict[:, :-1]
+
+            accept_mask = (torch.cumprod(accept_mask, dim=1)).sum(dim=1)
+            max_draft_len = self.retrive_index.shape[-1]
+            accept_index = torch.full(
+                (bs, max_draft_len), -1, dtype=torch.int32, device="cuda"
+            )
+            accept_length = torch.empty((bs,), dtype=torch.int, device="cuda")
+            extract_index = torch.full((bs * 2,), 0, dtype=torch.int, device="cuda")
+            eagle_verify_retrive[(bs,)](
+                self.retrive_index.contiguous(),
+                accept_mask.contiguous(),
+                self.retrive_cum_len,
+                accept_index,
+                accept_length,
+                extract_index,
+                max_draft_len,
+                self.draft_token_num,
+                triton.next_power_of_2(max_draft_len),
+            )
+        else:
+            # temp > 0
+            bs = self.retrive_index.shape[0]
+            predict_shape = list(logits_output.next_token_logits.shape)[:-1]
+            predict_shape[-1] += 1
+            target_logits = logits_output.next_token_logits[self.retrive_index]
+            predict = torch.full(predict_shape, -1, dtype=torch.int32, device="cuda")
+            accept_index = torch.full(
+                (bs, self.spec_steps + 1), -1, dtype=torch.int32, device="cuda"
+            )
+            accept_length = torch.empty((bs,), dtype=torch.int32, device="cuda")
+            expanded_temperature = batch.sampling_info.temperatures.unsqueeze(1)
+            target_probs = F.softmax(target_logits / expanded_temperature, dim=-1)
+            draft_probs = torch.full_like(
+                target_probs, 0, dtype=torch.float32, device="cuda"
+            )
+            coins = torch.rand_like(candidates, dtype=torch.float32, device="cuda")
+            tree_speculative_sampling_target_only(
+                predicts=predict,  # mutable
+                accept_index=accept_index,  # mutable
+                accept_token_num=accept_length,  # mutable
+                candidates=candidates.to(torch.int32),
+                retrive_index=self.retrive_index.to(torch.int32),
+                retrive_next_token=self.retrive_next_token.to(torch.int32),
+                retrive_next_sibling=self.retrive_next_sibling.to(torch.int32),
+                uniform_samples=coins,
+                target_probs=target_probs,
+                draft_probs=draft_probs,
+                deterministic=True,
+            )
 
         new_accept_index = []
         unfinished_index = []

sglang/srt/speculative/eagle_worker.py

@@ -185,6 +185,7 @@ class EAGLEWorker(TpModelWorker):
             self.topk,
             self.speculative_num_steps,
             self.server_args.speculative_num_draft_tokens,
+            batch.sampling_info.is_all_greedy,
         )
 
         # Free cache locations

sglang/version.py

@@ -1 +1 @@
-__version__ = "0.4.2.post2"
+__version__ = "0.4.2.post4"

{sglang-0.4.2.post2.dist-info → sglang-0.4.2.post4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: sglang
-Version: 0.4.2.post2
+Version: 0.4.2.post4
 Summary: SGLang is yet another fast serving framework for large language models and vision language models.
 License:                                  Apache License
                                    Version 2.0, January 2004
@@ -239,11 +239,11 @@ Requires-Dist: xgrammar>=0.1.10; extra == "runtime-common"
 Provides-Extra: srt
 Requires-Dist: sglang[runtime_common]; extra == "srt"
 Requires-Dist: cuda-python; extra == "srt"
-Requires-Dist: sgl-kernel>=0.0.3.post1; extra == "srt"
+Requires-Dist: sgl-kernel>=0.0.3.post3; extra == "srt"
 Requires-Dist: torch; extra == "srt"
-Requires-Dist: vllm==0.6.4.post1; extra == "srt"
+Requires-Dist: vllm<=0.7.2,>=0.6.4.post1; extra == "srt"
 Requires-Dist: flashinfer_python>=0.2.0.post2; extra == "srt"
-Requires-Dist: outlines<0.1.0,>=0.0.44; extra == "srt"
+Requires-Dist: outlines<=0.1.11,>=0.0.44; extra == "srt"
 Provides-Extra: srt-hip
 Requires-Dist: sglang[runtime_common]; extra == "srt-hip"
 Requires-Dist: torch; extra == "srt-hip"

{sglang-0.4.2.post2.dist-info → sglang-0.4.2.post4.dist-info}/RECORD

@@ -5,12 +5,12 @@ sglang/bench_offline_throughput.py,sha256=vIoF87HIpezB1x-xWzUl6SdXi88Fza8g4hDU7G
 sglang/bench_one_batch.py,sha256=d-LuRHEyDZjh180OCN5fqTjr8Zusk3zc0vhoJ33x0B0,17905
 sglang/bench_one_batch_server.py,sha256=iu73SsvYwnuRktYZDz1P6psMiRx8MbEbF5sbsYJdzYg,5962
 sglang/bench_serving.py,sha256=jYU3rYIDkzpYhjSpJw_IkEs_UNQfouNW4phs3z5TObc,54303
-sglang/check_env.py,sha256=4OqpZaEJOfBM6-vtPILto5kqDmgiZM1Koc7lK78A7CI,8427
+sglang/check_env.py,sha256=lDVA3ybt1wOE33HIMpkkU7zGRgLWez1_ifRRJ8qxbtw,8445
 sglang/global_config.py,sha256=fnT0U9vlHdGaQFKN9tYTnUF4-eVW4HYQURd5zvPtrg0,1286
 sglang/launch_server.py,sha256=mDXfwha8LHpWQJekcCosR98QhCQsbmilsBlI5jAIgg0,420
 sglang/llama3_eval.py,sha256=gWSboDchIGybIce88bJlrCG0yiLZ513mw4gcutJlzGM,10017
 sglang/utils.py,sha256=7HpOrPBhMivWH719m7Dy1rjrAXOAsnqelpwNBBbvjqs,13319
-sglang/version.py,sha256=64ujEkLGOA9yAnhsrnI5zBOk5lJIP4Z-b7gpnc9vbUo,28
+sglang/version.py,sha256=4aRfll4rDDWBVERpTNftyHYiEf_Z50L3jvLng1lO-j4,28
 sglang/lang/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 sglang/lang/chat_template.py,sha256=v4SyYViPHX3i3XT46F7vlARn4UaSiP3PBpTGtzO6uRY,17006
 sglang/lang/choices.py,sha256=-W1DVw9N9ZliVpvmWrzIXG4cswAah8eMQrHWzkS3D8o,6234
@@ -29,7 +29,7 @@ sglang/srt/_custom_ops.py,sha256=7jL5BTcoS8PmR56y2Qsa3q8emI-tmrJuV4hLTwLVFBE,504
 sglang/srt/aio_rwlock.py,sha256=6LYtOdeTUY3hkfa1dmYkgsaF2ttrwIF3hUWz2AZ2fqw,2970
 sglang/srt/conversation.py,sha256=USUoYiJf5DdHz7Ouclu30k3QSxMiem4WgZrA148MpSA,21695
 sglang/srt/custom_op.py,sha256=M5oqlgh32vAVeStFCruydTUfi_blGFJihVTnQBEOvwo,1134
-sglang/srt/function_call_parser.py,sha256=HMqCCd-YQeyADV_gTCduF9gmw2k3bRAkoJYcFsK3w3c,19230
+sglang/srt/function_call_parser.py,sha256=YmagXt1BIuTbeiWmSleZwJFCFR5r5EFqVQqKnJDYXiE,19568
 sglang/srt/hf_transformers_utils.py,sha256=_24uqCkZ4dvS9Uc5p2cCzX0Q8ShUzrh_Hp6mvg7hxHY,7729
 sglang/srt/mm_utils.py,sha256=1ScBunw_x4W8ebM_AcJ62-1T2mfT8NlMJqdAhkF1lb0,12367
 sglang/srt/model_parallel.py,sha256=eLXZhvJ4wG6dh0FontNCIdVZvHYdWgaeY-5cu7TD9tE,6078
@@ -46,7 +46,7 @@ sglang/srt/configs/load_config.py,sha256=la2ezNRcUZs7qiTYta2KEXqZ0U4TcmWW3U0sjoH
 sglang/srt/configs/model_config.py,sha256=sQIOfslBRzhOjucZdd8zE8nO9PEOc7zc6cZMbguQgoY,16876
 sglang/srt/configs/qwen2vl.py,sha256=ZjLy9v2eZY4wptUfY3CWgYKg2B5DDrkfCSyTy_Zf_bg,4351
 sglang/srt/constrained/base_grammar_backend.py,sha256=JFQFiAZLSqV6vck-ewIEzEEyncWLbRz_gkvkqpC282k,3185
-sglang/srt/constrained/outlines_backend.py,sha256=kNeJzr2NRqBy8AMiUlZUly0FmpR0raMB59QRhlfY_Tc,6970
+sglang/srt/constrained/outlines_backend.py,sha256=yPYgz44n-rSCStGGkS1lGazFiQzN7gqwSvpJ2YG0co4,7081
 sglang/srt/constrained/outlines_jump_forward.py,sha256=iZWXeR3gNYoMubLGyFmLPO4V2YsN5DiGjD71Xk9iFaE,6418
 sglang/srt/constrained/xgrammar_backend.py,sha256=l-37tdrPsp7xnxZpY8_0W1DnZSiBAH9e-BcwiAO8b0g,5048
 sglang/srt/distributed/__init__.py,sha256=jFOcyt-wFAPMBUAf9zkZalNQlt-4rqmT6pCKBz1E4qo,149
@@ -76,14 +76,14 @@ sglang/srt/layers/sampler.py,sha256=FIkh6sh91Fh5R8QJ6x66bJ8Y-xl5EfT4XVPXGXJ1l7I,
 sglang/srt/layers/torchao_utils.py,sha256=Ws24FdRBSkTpyeyA6bQrdDm-W5wfDxKvSIPUSahyMfA,4063
 sglang/srt/layers/vocab_parallel_embedding.py,sha256=txcjkuSDa6gZwESKj8X-HSLhAnMmDXL0FmFWY9SKqik,22155
 sglang/srt/layers/attention/__init__.py,sha256=KlQ0fl-o9v_NxBDhNZ4dPW2uQ2HeJjLm-0MTMWgaa28,2980
-sglang/srt/layers/attention/double_sparsity_backend.py,sha256=QEDF8tQKMkh-nbt4jHKHZhhgHuV0Fla_BPzzoo9JfT4,9231
-sglang/srt/layers/attention/flashinfer_backend.py,sha256=9BJEAQ5IcSMGvPfa6_D3cP9Gbo2XQ5GHBnF7cw2Rsng,42933
+sglang/srt/layers/attention/double_sparsity_backend.py,sha256=4mVyFPfZxPTwkQHGNCfI_4hQ8CbsWXJfxz-IQW77gAc,9143
+sglang/srt/layers/attention/flashinfer_backend.py,sha256=BlirMAu8dDicGdpZiW51hx9FpiU3EY-MNhZ1vdCJxGo,42398
 sglang/srt/layers/attention/torch_native_backend.py,sha256=KrcAqTLVZLtwgOmB0xhwUUsX32M-5LYZpNxaRNT4VuA,9252
-sglang/srt/layers/attention/triton_backend.py,sha256=PZU496wPzyRopA600riR5sxz-gIz8u9TYz0MzmMuX5Y,8858
+sglang/srt/layers/attention/triton_backend.py,sha256=mtLs768rhtCF_BVAV_rmYac0U4R1_HHc-9ic4JratsY,10100
 sglang/srt/layers/attention/vision.py,sha256=zLjKmzUlkgq1RFcP3b4EPArOAKovoaDLgYfM5SyB2wM,13181
 sglang/srt/layers/attention/triton_ops/decode_attention.py,sha256=tcUAdacBWTpZmro7vZeRPasfwRWFlCR4bxfGpFOYgZ8,17831
-sglang/srt/layers/attention/triton_ops/double_sparsity_attention.py,sha256=1pSXfY3EEaM7iRN_uElHnAfsrJMhTFbu9fj8Z0O2PbE,21480
-sglang/srt/layers/attention/triton_ops/extend_attention.py,sha256=DWOZXSTVN5ZbcFjDjcqs-nPdUkxSwum0SVXhVKqwh2g,11688
+sglang/srt/layers/attention/triton_ops/double_sparsity_attention.py,sha256=ztLWKeW-260EiIw3kCAbtUTUHHxAICz2mVxZJFes4oI,31167
+sglang/srt/layers/attention/triton_ops/extend_attention.py,sha256=R6QgrcBf6XuLzQ1jamrILNypaPi3ynkMPTfjae0d3JA,12695
 sglang/srt/layers/attention/triton_ops/prefill_attention.py,sha256=Y66gZ37u0GKMPtI8n5MbO6uOxRuGEmKIG0IPbJTOqAM,6213
 sglang/srt/layers/moe/fused_moe_native.py,sha256=OEWpM93X5tJG4-rwz5qmdpTzEUR73zun29YRV3bZglY,4269
 sglang/srt/layers/moe/topk.py,sha256=6A4W1ztlV2dQvkXcPJvFvAg0QEhE58Q7eE7iw8N36J4,7230
@@ -91,7 +91,7 @@ sglang/srt/layers/moe/ep_moe/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NM
 sglang/srt/layers/moe/ep_moe/kernels.py,sha256=wb_S2qLxoWWgQu9coXy0XLNGvHzdZSdwXr0PGy4QySg,10940
 sglang/srt/layers/moe/ep_moe/layer.py,sha256=aS8t1XUvlTnO9IQaxGjW5bOXP4FrJDXzymEIvlIDMro,22603
 sglang/srt/layers/moe/fused_moe_triton/__init__.py,sha256=h9yMFAL_bagUf-qBED8gSWdCOb7d8IdA-pE-L_nIg8E,842
-sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=_uUn0EKf_trPyKP4jfCytKKgnC3ziUM_X9L_PbaXhbE,37241
+sglang/srt/layers/moe/fused_moe_triton/fused_moe.py,sha256=NckXo7tQ0rp3trdAq5TB5mP4MmCdjm6RIN157d9Efvk,37720
 sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=-49WRpq9OtRZocQjW-YNcB_ruK09nIJqGHKNa8CJsws,22691
 "sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB,dtype=int8_w8a16.json",sha256=iNGsE2ZeVnQEnN4A8UJ9Jv0d3hbRF2MJ9oBgjup5Szk,2737
 "sglang/srt/layers/moe/fused_moe_triton/configs/E=1,N=14336,device_name=NVIDIA_A100-SXM4-80GB.json",sha256=JJN0hryyLr5Zv3dSS7C8cPFhAwTT6XxUVnBGMZvV6JA,2752
@@ -196,24 +196,30 @@ sglang/srt/layers/moe/fused_moe_triton/layer.py,sha256=-49WRpq9OtRZocQjW-YNcB_ru
 sglang/srt/layers/quantization/__init__.py,sha256=_Sba1KQnmZNKGDKM1MfBs2T3uDqOHfeW6IHO2mTUvfs,4471
 sglang/srt/layers/quantization/base_config.py,sha256=daK9p0aijMszLUm1W4Pc33FK87MdqYK1NoWFKif-j80,4599
 sglang/srt/layers/quantization/fp8.py,sha256=ibttPVCUsCQ0LXy7FUb8wnzqGcGZQXQLqwCB4a2fai4,35160
-sglang/srt/layers/quantization/fp8_kernel.py,sha256=gm_mDHeBswggsaNMEo0niayx9DS2mGJkLkrnnOYQnlU,15709
+sglang/srt/layers/quantization/fp8_kernel.py,sha256=qlXXT9WO3TKxZv9r0rAdcDjO_jZYobtKnvTxmHDrfoo,16304
 sglang/srt/layers/quantization/fp8_utils.py,sha256=7v-RNwuYXa-gPO3msRDB0Z3uajOQMYd2Cj0NMoq1hg4,4148
 sglang/srt/layers/quantization/int8_kernel.py,sha256=t_BLVf8XjOyn7S3Lu3B4hXvw8DvTg4Anco7TNadL58U,1436
 sglang/srt/layers/quantization/modelopt_quant.py,sha256=_VdVz77dTP-IczPeFrdH6Ttro2D26BZvMlZkCKWj_5o,6200
 sglang/srt/layers/quantization/w8a8_int8.py,sha256=PBapNo27rkUdX95E0ihUZeHTXP2Id3CwlNlMU8GIuZc,3344
 "sglang/srt/layers/quantization/configs/N=1536,K=1536,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=tkLjwLC_aVXhzuvo-2QHkojXZauPJsf3jNHFn1S7uRA,3244
+"sglang/srt/layers/quantization/configs/N=1536,K=1536,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=CPo1WRF0HgsQMPBkvpoImElQMrfwpJLhEvL86e6fkPU,3247
 "sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=xqJNygFgPCe4jDpkfVOWCXpuNMUjmssdD8oGhp2iXv8,3726
 "sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=xqJNygFgPCe4jDpkfVOWCXpuNMUjmssdD8oGhp2iXv8,3726
 "sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=Qoj9rLLRDbKM4IKBCXvN8RcxzSmNPd0TQUiM7CXDqHI,3241
+"sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=UZljnxxCSjwnZlX3OgKWZJGXCf5BWF_agEpNX8I4Zxc,3248
 "sglang/srt/layers/quantization/configs/N=1536,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=4D3Ku4y7BCVEJzueKvQC_KvOR026w3ONWsxfsA_YrEc,3249
 "sglang/srt/layers/quantization/configs/N=2048,K=512,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=7v4tp0RaT4vxF4urSBrkK5FR_5ikeFQ1htF3DwDl1lk,3249
+"sglang/srt/layers/quantization/configs/N=2048,K=512,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=3matoCzEe4aexwoe7YTmkjyE4NA8khWXjL5EySuNwzA,3254
 "sglang/srt/layers/quantization/configs/N=2048,K=512,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=0J2MFgaLkv-mfVE5x363lgVKYU6miLG_xRO3tJUga_M,3249
 "sglang/srt/layers/quantization/configs/N=2304,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=4ubbhwSFX_XbefRLEkLoWxJkcetFWPzsszPu0X3_Wrw,3242
+"sglang/srt/layers/quantization/configs/N=2304,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=9f8Ib4gLEFSfdNpO8IL8uiONImvqnlPbJrZ0HM3OB-o,3247
 "sglang/srt/layers/quantization/configs/N=2304,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=FhyniGTx5QeCuVrBSVTQys6q05Pr5lPEcPykpAX7Iyo,3247
 "sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=0v17v78pETXv6S2ZoibekxOVhiTmCm807DYG4DONUck,3259
 "sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=I44PvJj758-sw_fCOVROLTpG0NQ5_5PCYyQcpZC1YSY,3259
+"sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=tG5_iVeRBHTgHX-liOf79nWRjj_lUZ-NQWTbBrBgORQ,3246
 "sglang/srt/layers/quantization/configs/N=24576,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=-CVHqClROli9FWe_FnlnuAG2LiFivDFK_nghH6t-BWc,3261
 "sglang/srt/layers/quantization/configs/N=256,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=GsLoYkaZ2p4Qu0Coj-X90s7JWyfZBOloIHPlyNKSIes,3246
+"sglang/srt/layers/quantization/configs/N=256,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=BpKweY4HfBx5xvqSvwNahy9x7R5FH-YK8j6rLFQdKwo,3248
 "sglang/srt/layers/quantization/configs/N=3072,K=1536,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=I6a5wQ0R3cBp8ix_PDamWZN2aJmV_1p0tg430L3Updg,3727
 "sglang/srt/layers/quantization/configs/N=3072,K=1536,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=I6a5wQ0R3cBp8ix_PDamWZN2aJmV_1p0tg430L3Updg,3727
 "sglang/srt/layers/quantization/configs/N=3072,K=1536,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=8zuJhFdd6aXREpiqPFhIKEFWA5lgLVGrG0-a9UXcBqk,3262
@@ -223,6 +229,7 @@ sglang/srt/layers/quantization/w8a8_int8.py,sha256=PBapNo27rkUdX95E0ihUZeHTXP2Id
 "sglang/srt/layers/quantization/configs/N=3072,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=M3nwpZd2-0w263ywZt9gaw53z7MN673T5tl4tc43Ntk,3249
 "sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=vLoV3JMtvHOKpR5D1BeCQPMuYlWUAlrXu54gByNkwKY,3266
 "sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=Mtw7a9BSspj2TzC-aPxE82o1LEvwzgbUuIofwRxUNA0,3263
+"sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=B0lo3SuoQXhBEnojH2TwpVeurvlKD8yI8kQrJ5ORhWU,3249
 "sglang/srt/layers/quantization/configs/N=32768,K=512,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=NHdx3tZnfLF7NplswMzcTRbQEQFLtChg4rd7GU9lMbM,3262
 "sglang/srt/layers/quantization/configs/N=36864,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=laYeH4w0iZOj2Yg3vDgtKoroNQnwBEX4GUGLrO9095I,3260
 "sglang/srt/layers/quantization/configs/N=36864,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=TWcPDZ2miQMD6OWDC1FteRs80ND9RC-oJL3PLVmJbtI,3257
@@ -244,17 +251,23 @@ sglang/srt/layers/quantization/w8a8_int8.py,sha256=PBapNo27rkUdX95E0ihUZeHTXP2Id
 "sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=xsFMrq4aybClfJyhm78c1Hf1jcyFSGnfygdHYp7OhSQ,3727
 "sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=qG6v3n3qF6LE2DdGT-mDIXecZ1a7vg7p3QqXYCMX85k,3254
 "sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=EgFTGyW_YuDwyEDUCoGglyI1ETdj9J7AR0UfJ86jMoI,3249
+"sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=BpKweY4HfBx5xvqSvwNahy9x7R5FH-YK8j6rLFQdKwo,3248
 "sglang/srt/layers/quantization/configs/N=576,K=7168,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=NiorJgOotxkQcP49ID3z5al1UA4QQDrT8MvbCwAWL5Y,3248
 "sglang/srt/layers/quantization/configs/N=7168,K=1024,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=bPQWtvaJrzOOIgI-R-MIxs_f4yC_FobkDydu3OkOFtg,3252
+"sglang/srt/layers/quantization/configs/N=7168,K=1024,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=h32lCTFilLlyKbMeuJvNWG1v0yJJzNj93kwSvlrHfaY,3249
 "sglang/srt/layers/quantization/configs/N=7168,K=1024,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=ZRgiuHZ2SFC6u-WV5DGwau4k1RiPLI67eENO0e-5Ylg,3253
 "sglang/srt/layers/quantization/configs/N=7168,K=1152,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=-hP_P8NM0K04mGzTmpGBNibQ5xxh5gPz5WtoMXhoz1E,3253
+"sglang/srt/layers/quantization/configs/N=7168,K=1152,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=0jX-z2lTgVw7ABLmWsIsQdqW4EjmbXKRDHye_XPLCAE,3245
 "sglang/srt/layers/quantization/configs/N=7168,K=1152,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=FB5Le4obvPoCgFSnC_3-Uh59n-Mt4Rol8saXVcK3RPw,3252
 "sglang/srt/layers/quantization/configs/N=7168,K=128,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=kLviGvVngpgOuelfKtvv9Is7MWQ89rGxlomMRP6t0Ic,3250
+"sglang/srt/layers/quantization/configs/N=7168,K=128,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=_exM3wJ3FMmGHweBcH-8IxwZBzaOmPaF3ScMM6KDpiY,3253
 "sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=vlys0Zi_CaaU41OHGbWSBtbVglFi98bgqEySBMc9Sdg,3258
 "sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=YWyByOlKSqp5lbcUa8eu6N2dHRKJqJDbCDSjdDQJngg,3249
+"sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=ENRWYdUwI0ooHb6IwcHliupRWOPnw-7-WtxZB-qQGJI,3245
 "sglang/srt/layers/quantization/configs/N=7168,K=16384,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=j5PTW0IC4Z2yQIygcdICaOsvb639u6Mv-ZpJYkrBQ2k,3254
 "sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=Ggy4hejkcWjiw5Bi-wGzSP5JLVuvOjip_rbjXFBJZbs,3257
 "sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_H100_80GB_HBM3,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=Xy4mgZx5iiEvuv2ydO4dFNIT8s0jgBhNHE1vu93fGJM,3250
+"sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_H20,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=TdWuE2RIsIyr4Im24MuWK3XyiNtbhO_hAiAXDz5gNUk,3246
 "sglang/srt/layers/quantization/configs/N=7168,K=18432,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=x476nFeltB_2iO9_6y-z2P_unAbh7ghLPFi5z2LOTOo,3253
 "sglang/srt/layers/quantization/configs/N=7168,K=2048,device_name=AMD_Instinct_MI300X,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=sVbH4YRLTxBqvTh_6xbtXkj3orOrKytlwM-_4gtD6IY,3725
 "sglang/srt/layers/quantization/configs/N=7168,K=2048,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=sVbH4YRLTxBqvTh_6xbtXkj3orOrKytlwM-_4gtD6IY,3725
@@ -270,17 +283,21 @@ sglang/srt/layers/quantization/w8a8_int8.py,sha256=PBapNo27rkUdX95E0ihUZeHTXP2Id
 "sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=AMD_Radeon_Graphics,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=aoq4792zPo87QO7VrEf9fb_vj0zPiHIu7Ho9aMXwcLw,3731
 "sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=NVIDIA_B200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=PD4AJYCkHfy2ivv9baMouFXzBTy0eKMumbAfxfm91HI,3256
 "sglang/srt/layers/quantization/configs/N=7168,K=256,device_name=NVIDIA_H200,dtype=fp8_w8a8,block_shape=[128, 128].json",sha256=FFBjSWlpKXMxfAUUYUqXbOK_Hd7qBeBsfbcaa9uB4qY,3249
-sglang/srt/lora/lora.py,sha256=71sVFPB23THA4O06v2_G91vaY3MhAchsti-LErycJ6g,14869
+sglang/srt/lora/layers.py,sha256=r34oprzwyE3SWPvaNkBvXWPtfa-0IY987_bjj36ySfw,9996
+sglang/srt/lora/lora.py,sha256=_WrZxS6-sarwUPvumcReyKGrH6fSCd8-UsoX56aQJ4s,7293
 sglang/srt/lora/lora_config.py,sha256=a2fTQESlCbG1xLiBYy4ptZ6c0Burcqyg1_6V1XSok-Y,1506
-sglang/srt/lora/lora_manager.py,sha256=fa-Sy4b3k_YzJJY5XQgxCX5AycnraF-f600jDBaXKzo,13991
-sglang/srt/lora/backend/__init__.py,sha256=ul5yTkHrwDTv8RZXu_Miw5_XpBlxqKeGmLfUT6Ph7EE,211
-sglang/srt/lora/backend/base_backend.py,sha256=MZfUhIwujwDyN2hFoVMB3N_gJokel6aU4Scc8fv5_L0,3575
-sglang/srt/lora/backend/flashinfer_backend.py,sha256=qXqtt4w_-Gqkk-j27CHHI1Jeu0gb8jr0ngI54eCNdqI,2845
-sglang/srt/lora/backend/triton_backend.py,sha256=MzVZYtPXwqS8DfeFMIpI0xNI90oVHcasaT3ppea9Vns,1722
-sglang/srt/lora/triton_ops/__init__.py,sha256=F4NrXZ4bSD39hkPVkby6OR-0Ap7dOOUsB0BjqjGMjy4,195
-sglang/srt/lora/triton_ops/qkv_lora_b.py,sha256=gFO53wGm93GpDmFaJgqyD1tgmrDllZ2bCBhGczZg0_4,5664
-sglang/srt/lora/triton_ops/sgemm_lora_a.py,sha256=GvPFXGXC2qLh4djlIIyAvobo9VLJcSEtyjcaveQh0Cw,3981
-sglang/srt/lora/triton_ops/sgemm_lora_b.py,sha256=pZxVJOMaf0OwMUJAw37KA4ZcoQb8gLpFGbYPirX8Oqk,4298
+sglang/srt/lora/lora_manager.py,sha256=-7ZWAL-E2mW3acXd9M0Z_slnTV1GfzGBD4RRv3cjafs,7851
+sglang/srt/lora/mem_pool.py,sha256=eV_GXETxNODPVIAnTEeUUUVn0IVgguBR_mYFzIK-VHA,6835
+sglang/srt/lora/utils.py,sha256=6i7Q1Y-1LLbRkeCMv_lKIzkTN0veUTLbc8wlHn7R-bA,4571
+sglang/srt/lora/backend/__init__.py,sha256=98L_KRRnE3gcGcx7Lb6yjAEUUE_Yay3QszcQXdzYsDw,708
+sglang/srt/lora/backend/base_backend.py,sha256=dldwA7vTWrB1ln1MwLYKNtMkBoAgD7OLSlWe9tL2lzk,4602
+sglang/srt/lora/backend/flashinfer_backend.py,sha256=fXfkl7Cpw8ap2bCrgWdn_gEUzMXX1pNjNuiPw3kA76U,3984
+sglang/srt/lora/backend/triton_backend.py,sha256=ZT5M30vj8x77Kltukpga4wk1sd8fT4n_FdsOMQBTMI0,2610
+sglang/srt/lora/triton_ops/__init__.py,sha256=JGOYPIn1XbGcyJTbt8A0qoc02PYONSGNNjGkC8yJpAM,283
+sglang/srt/lora/triton_ops/gate_up_lora_b.py,sha256=qve4oNZHYUFk9ckmT2BVuDNMEvrN7Quu6RsS8Iz3uRQ,5066
+sglang/srt/lora/triton_ops/qkv_lora_b.py,sha256=BmIcTZMnlSnie9rnMl4KvLpc4Njsk7_IppbUqitf9Xw,5738
+sglang/srt/lora/triton_ops/sgemm_lora_a.py,sha256=kv-AvJ_Bi3yWjGvFnSwXvP66iJvY9n9pEnJzJ9-DWzo,3982
+sglang/srt/lora/triton_ops/sgemm_lora_b.py,sha256=Ai5vPriT4OgACwK7xrpGgf5L1oaN9x0jwNKMChu3uI0,4299
 sglang/srt/managers/cache_controller.py,sha256=DXnIunJgtTws1WF2vZOYVQe56vacV7Mn4wL9zoG8Xz8,10909
 sglang/srt/managers/configure_logging.py,sha256=aY9xExurz7t_IdItd-9GuVuM7kEGB8_bRryhZxKdu9o,1542
 sglang/srt/managers/data_parallel_controller.py,sha256=b64aC6iLr5RolJyNQnT-yTQ_TSI9DDLtuABf_TPTUrM,9421
@@ -302,7 +319,7 @@ sglang/srt/mem_cache/memory_pool.py,sha256=9ud97u1cXnN6O0qlR8tv8woN_20gqisTV6aBg
 sglang/srt/mem_cache/radix_cache.py,sha256=hVILXvc5PauHuLTeyZbm3NCf3AOimaAuXjll53MSLeU,11754
 sglang/srt/metrics/collector.py,sha256=_yl0_paSARxS1ypZgd-pLJ29tMizolHuwROX21dOXTk,7326
 sglang/srt/metrics/func_timer.py,sha256=VFyNRrbnKVCwnQsrlLin1lITJfjQpf9m8sGPqL5LIsQ,3438
-sglang/srt/model_executor/cuda_graph_runner.py,sha256=Tv9ktSMaHNcAUUU2ERmxouFc6c94qbNMGt7_usQCD3w,18202
+sglang/srt/model_executor/cuda_graph_runner.py,sha256=7MLskhry5KFP-lhqUfAwZ8P4HckFCEQonYqIDjfAWSw,18455
 sglang/srt/model_executor/forward_batch_info.py,sha256=t1RlBgoeS-_Ikl28Xjvt-aouh1nNUc3eLM4iGY4_QqY,14988
 sglang/srt/model_executor/model_runner.py,sha256=5ET8CXwSrzIKb_rDW-6S0p7tv-iCPoEwZk3dESwPdF8,33090
 sglang/srt/model_loader/__init__.py,sha256=zGZkOBz1zx-pkaIy47BasL3fjDlAcxAXUTjInOhXHAE,919
@@ -325,7 +342,7 @@ sglang/srt/models/granite.py,sha256=3HqQXJlfoKd11w1NCpTYmiPO9HlkA1jJqoAmuTzHuU0,
 sglang/srt/models/grok.py,sha256=NXC0I5_wXmlQ0-gMWgiT-X9ebzOsrTJGcltAXkY6064,18030
 sglang/srt/models/internlm2.py,sha256=INGGwSCYKoZRAokXJC78RKKde2fgHn9P4JG-N37Pfn0,12124
 sglang/srt/models/internlm2_reward.py,sha256=8K26A9oIFFGx_9U2mF87j7FX8K87HGKMnVL3ht1Uc7I,2398
-sglang/srt/models/llama.py,sha256=YKtLpxgk_OmPRpBJSr1BCTWC6IILFzZtakKnWYYHKw0,22040
+sglang/srt/models/llama.py,sha256=hGBUo-70o0vLT6BI-v32qEv_g8Vr7ItEpqxAt1Mf9-0,22248
 sglang/srt/models/llama_classification.py,sha256=DwboM1xHXdf3Fddf7xGnrfdOLJwXdiJs994cIpAPa2g,2984
 sglang/srt/models/llama_eagle.py,sha256=88DzR54DKBIKJ1h-bkIa8mc1qJnlkdZ1eGYY3c5mpBY,4442
 sglang/srt/models/llama_embedding.py,sha256=rh-AiczPY_pTpzcACHvSMVjh1hsV_MZBBwP0LQxPsGM,3130
@@ -365,10 +382,10 @@ sglang/srt/sampling/penaltylib/penalizers/frequency_penalty.py,sha256=1Zp2aL6dD6
 sglang/srt/sampling/penaltylib/penalizers/min_new_tokens.py,sha256=_Nxv0XgUPirZjw2SEJYp_Cd9ZcLwmt7h6JE6J4hhFq4,3629
 sglang/srt/sampling/penaltylib/penalizers/presence_penalty.py,sha256=5tOgCg7OvE9kSN9VMCpH1hwqo1YMxt9iS5PVpct9HpU,2468
 sglang/srt/sampling/penaltylib/penalizers/repetition_penalty.py,sha256=l1DyU8kC8n_F4Z6Jd8mZKfF23buuLZ5dWuVfyqDWkUI,2968
-sglang/srt/speculative/build_eagle_tree.py,sha256=GlHDIbaZInhf1LYuWVmkVCvPcd6sKDxeXafLTai7Zpw,9892
-sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=5ZCy6ndPA2p95xDgo2kXWD3zCtVaq4q5X0HBpAbB3Xs,7929
-sglang/srt/speculative/eagle_utils.py,sha256=DUANfRddoQ4MqNuQW3Uz21qWRaetYf4prVcljZR0tyI,22945
-sglang/srt/speculative/eagle_worker.py,sha256=4oROLwUBJIwEHNHNEfvsy74DqLQLVc4KfjdR-MrB1OM,12038
+sglang/srt/speculative/build_eagle_tree.py,sha256=zWthboIgzPzSOXcGxDpDv0rBOQP55HYGrBKGqm2gWF0,20732
+sglang/srt/speculative/eagle_draft_cuda_graph_runner.py,sha256=FY4hcwd0Blx7AXbeX6quaXPNgWA8WGIqVcQiEgHyERk,8002
+sglang/srt/speculative/eagle_utils.py,sha256=ypjVmVTVzCGclOVHRMJxdLUSPkf1-7bNXQS0oP6dn5U,25644
+sglang/srt/speculative/eagle_worker.py,sha256=yDCU4liKU_jgtwQ-8L9_GuUtFRo3l1Df7mCeKX8yHCo,12085
 sglang/srt/speculative/spec_info.py,sha256=D7A27UU1iOwIBEjXTgAxZ7jdftbTiVlMCvK8GmYr2zg,488
 sglang/test/few_shot_gsm8k.py,sha256=7yDbEQe49gZeJhz2wFFX-gf_59ThDKsCS1xwfogNc7k,4034
 sglang/test/few_shot_gsm8k_engine.py,sha256=QQbrwOX6-cJDD3RZC_e7zPnt6aSo8JdF8X_lRHSjdDM,3886
@@ -386,8 +403,8 @@ sglang/test/test_layernorm.py,sha256=IacByD5d-stXjzBz8Ypamc7povlcedpKPbb_4JLgo3c
 sglang/test/test_programs.py,sha256=aUV9Ex_B714ph7ytv6W3J7sdGDKC6lGIhUy95Yg6AHQ,18878
 sglang/test/test_utils.py,sha256=BU6lAX3bu3TNQZqVC9UPnyq3I7iV5kigHQKJx7UNlOQ,26192
 sglang/test/srt/sampling/penaltylib/utils.py,sha256=CjxHgywh0hx_87iynzQt_ztHu6zBVuE-YrZ-XPmW6U4,12906
-sglang-0.4.2.post2.dist-info/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
-sglang-0.4.2.post2.dist-info/METADATA,sha256=-Cl1_HFESAZxOXkBnaqDshP2M93b_4FWYGVh-1Yyw3s,23763
-sglang-0.4.2.post2.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
-sglang-0.4.2.post2.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
-sglang-0.4.2.post2.dist-info/RECORD,,
+sglang-0.4.2.post4.dist-info/LICENSE,sha256=FJXh51fvTQklojUFY89XVLsjxRcBqOxPs8XNy-2uZ0c,11346
+sglang-0.4.2.post4.dist-info/METADATA,sha256=VylN7wgaeE_8T-ddwxTmujJo04JCRpaP_uyhGfo-InQ,23773
+sglang-0.4.2.post4.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
+sglang-0.4.2.post4.dist-info/top_level.txt,sha256=yxhh3pYQkcnA7v3Bg889C2jZhvtJdEincysO7PEB09M,7
+sglang-0.4.2.post4.dist-info/RECORD,,