Commit Graph

  • ee6000f656 Update CI workflow to exclude feature branches (#4960) plusNew001 2025-11-12 14:44:43 +08:00
  • bbd9c96ab9 [XPU] [CI]Change CI to multi-concurrency (#4924) plusNew001 2025-11-12 14:43:55 +08:00
  • 76e60e98f8 [Iluvatar][CI] fix safetensors_rust.SafetensorError: framework paddle is invalid (#4972) yzwu 2025-11-12 14:13:40 +08:00
  • 35bd2afab3 [Benchmark] Add GEMM & MoE kernel bench (#4809) Sunny-bot1 2025-11-12 11:56:40 +08:00
  • 8a96944a0a [CI] Update PORT range to avoid conflict with system ports (#4953) YuBaoku 2025-11-12 11:17:49 +08:00
  • 09cd6c5d3e Modify README Jiang-Jia-Jun 2025-11-12 11:03:23 +08:00
  • 9c52d9eb8f [CI] remove useless tests in docker_build (#4974) YuBaoku 2025-11-12 10:55:09 +08:00
  • ff653503ff [Docs] Add License in Unittest (#4957) Echo-Nie 2025-11-12 10:44:09 +08:00
  • 2aabaecbc2 [CI] Add five unittest (#4958) Echo-Nie 2025-11-12 10:43:33 +08:00
  • a5103eb198 [CI][XPU] Change Paddle Version to Nightly (#4973) plusNew001 2025-11-12 10:29:16 +08:00
  • b09ebb2813 refactor pt loading (#4532) bukejiyu 2025-11-11 21:30:39 +08:00
  • 4c911ecb74 [CI] fix apt_sources error of focal in docker_build (#4961) YuBaoku 2025-11-11 20:35:06 +08:00
  • f20f29fc79 [CI][XPU]Update health check endpoint to use port variable (#4965) plusNew001 2025-11-11 20:19:53 +08:00
  • da6b4c10e5 [ATTENTION] make buffer alloc as a function (#4945) 周周周 2025-11-11 19:17:08 +08:00
  • 08b96baa4a [Iluvatar][Doc] Add ERNIE-4.5-VL-28B-A3B-Thinking doc (#4955) yzwu 2025-11-11 19:15:19 +08:00
  • 896ef565cc [Others] Add Tests for GPU Model Runner and Logprobs Output (#4913) chen 2025-11-11 18:37:33 +08:00
  • a83250ae3f [CI] Update test_api_key.py (#4948) kxz2002 2025-11-11 16:49:54 +08:00
  • 76be598129 replace paddle.max by numpy to avoid useless error log (#4893) K11OntheBoat 2025-11-11 16:28:05 +08:00
  • 3098aee05f [Perf] Support tensor transmission between work and engine with zero-copy to improve efficiency (#4839) SunLei 2025-11-11 15:43:11 +08:00
  • 8b61f01c68 [CI][XPU]Update run_ci_xpu.sh to lock paddlepaddle-xpu version (#4949) plusNew001 2025-11-11 15:38:05 +08:00
  • 38ccf9b00b [Docs] release docs 2.3 (#4951) ming1753 2025-11-11 15:30:11 +08:00
  • 5280b9e0b4 [XPU] fix xpu deployment md (#4941) Lucas 2025-11-11 14:39:52 +08:00
  • 215cda2f80 [XPU][Doc]Update XPU release2.3 note (#4939) yinwei 2025-11-11 11:57:49 +08:00
  • 3f09ebf3da Update model names in FastDeploy v2.3 release notes Jiang-Jia-Jun 2025-11-11 11:53:26 +08:00
  • 75294bcfb1 [Docs] add ERNIE-4.5-VL-28B-A3B-Thinking instruction (#4944) LiqinruiG 2025-11-11 11:40:52 +08:00
  • c0a4e2b63b Update README.md Jiang-Jia-Jun 2025-11-11 11:38:30 +08:00
  • 7bedf2041a Update README.md Jiang-Jia-Jun 2025-11-11 11:37:31 +08:00
  • 3707af7a4f [Iluvatar] add vl into ci and support v1 loader (#4774) yzwu 2025-11-11 10:50:17 +08:00
  • 07a82afcae add tie_word_embeddings for lmhead (#4916) Ryan 2025-11-11 10:46:35 +08:00
  • 3f74281496 [Docs] add ERNIE-4.5-VL-28B-A3B-Thinking instruction (#4937) LiqinruiG 2025-11-11 10:43:44 +08:00
  • d7f14dba8b uodate docx (#4938) yangjianfengo1 2025-11-11 10:28:46 +08:00
  • 3dc0ffa46d [TSP] Support qwen3 moe tsp + cudagraph (#4871) Yuanle Liu 2025-11-10 23:37:51 +08:00
  • fb2eb403ab [Opti] Unlimit zmq message lens limit (#4465) chenjian 2025-11-10 21:38:02 +08:00
  • 927bd74075 [Docs] add doc for glm (#4933) chen 2025-11-10 21:21:33 +08:00
  • cba7b2912f [Opti] Unlimit zmq message lens limit (#4934) v2.3.0 chenjian 2025-11-10 21:16:36 +08:00
  • 197a0f7af4 [BugFix] fix VL fp8 bug when moe token_num is 0 (#4929) ming1753 2025-11-10 21:16:10 +08:00
  • 3665c283b5 [XPU] [CI]Change CI to multi-concurrency (#4866) plusNew001 2025-11-10 21:09:48 +08:00
  • 0a6981f928 [BugFix] Fix inference_start_time (#4922) (#4930) kxz2002 2025-11-10 21:07:52 +08:00
  • 59d2edde29 [BugFix] Add support for weight shape constraints and group size selection in Machete (#4911) Sunny-bot1 2025-11-10 20:57:35 +08:00
  • 2926bf60f7 [XPU][Doc] Update XPU release/2.3.0 documents (#4931) yinwei 2025-11-10 20:07:34 +08:00
  • f7159e31ba [BugFix] When the value of "temperature" is 0, adjust it to 1e-06 (#4919) luukunn 2025-11-10 19:34:20 +08:00
  • 2dfbcf3cc9 [BugFix] Fix inference_start_time (#4922) kxz2002 2025-11-10 19:28:44 +08:00
  • aa79e6185a [Docs] Improve reasoning_out docs (#4901) LiqinruiG 2025-11-10 19:20:38 +08:00
  • 07b21d241d [XPU]Update documentation (#4917) qw86972190 2025-11-10 19:11:42 +08:00
  • 54536267db [DeepEP] support P async_finish (#4899) 周周周 2025-11-10 18:24:02 +08:00
  • 78895e2c7d [Bug Fix] fix bug for PD EP (#4823) chenjian 2025-11-10 15:33:29 +08:00
  • 4125b97603 [Fix] Fix eplb for ep mixed (#4894) xiaoxiaohehe001 2025-11-10 14:46:26 +08:00
  • 112623e33e init version, exist some bugs, waiting fix (#4906) Echo-Nie 2025-11-10 14:16:09 +08:00
  • 90b0936ae9 [Docs] add api-key usage instructions (#4902) LiqinruiG 2025-11-10 13:39:39 +08:00
  • 41c0bef964 [BugFix] When the value of "temperature" is 0, adjust it to 1e-06 (#4900) luukunn 2025-11-10 13:24:33 +08:00
  • c6e9717f33 [XPU][CI]Update test assertion and base response value (#4908) plusNew001 2025-11-10 12:59:04 +08:00
  • 0a3bc84f71 [XPU][CI]Update test assertion and base response value (#4907) plusNew001 2025-11-10 11:44:54 +08:00
  • 8a9e7b53af [Docs]Supplement the English and Chinese user documentation for Tool calling (#4895) zhuzixuan 2025-11-08 20:05:14 +08:00
  • 87911b7cf1 [Feature] Enable FastDeploy to support adding the “--api-key” authentication parameter. (#4806) kxz2002 2025-11-08 18:24:02 +08:00
  • 80aedb82ce [BugFix] max_lgprobes=-1 maps to ori_vocab_size (#4884) chen 2025-11-07 22:15:40 +08:00
  • 6de1ce3b25 [Metax] support ERNIE-4.5-VL-28B (#4820) Neil Zhu 2025-11-07 20:55:49 +08:00
  • 69e503499d [CI] fix docker_build error of ciuse (#4886) YuBaoku 2025-11-07 19:44:21 +08:00
  • 6871aad03d [BugFix] fix token_processor zmq (#4827) chen 2025-11-07 19:43:25 +08:00
  • a7ef998e04 [Feature] Optim PaddleOCR-VL (#4872) ming1753 2025-11-07 17:55:02 +08:00
  • fa098383f6 [XPU][CI] Ci bug fix (#4889) plusNew001 2025-11-07 17:50:11 +08:00
  • 329e999f2d [XPU][CI] Release ci bug fix (#4892) plusNew001 2025-11-07 17:49:41 +08:00
  • 51ef990e21 [XPU] modify 424B model deployment parameter (#4887) ddchenhao66 2025-11-07 17:34:49 +08:00
  • 72d5ee9a7c [XPU] modify 424B model deployment parameter (#4888) ddchenhao66 2025-11-07 17:34:37 +08:00
  • 6b5ae9ffea [XPU] fix ep_tp all2all ci (#4876) zhupengyang 2025-11-07 16:37:23 +08:00
  • 3b0bdbae65 [XPU] fix text_image_gather_scatter when image_token_num == token_num && text_token_num == 1 (#4881) Lucas 2025-11-07 16:35:49 +08:00
  • 3dbe5596e6 [Feature] Support eplb for ep (#4786) kevin 2025-11-07 15:42:29 +08:00
  • cba185f1fe [Feature] Optim PaddleOCR-VL (#4873) ming1753 2025-11-07 14:56:44 +08:00
  • bbe0820555 Add instructions for copilot reviewer Jiang-Jia-Jun 2025-11-07 11:19:27 +08:00
  • 79e6bf4bdc [Others] Delete PaddleOCR Useless Function (#4815) Haonan Luo 2025-11-07 11:14:41 +08:00
  • 048856a7f6 Add instructions for copilot reviewer Jiang-Jia-Jun 2025-11-07 11:01:05 +08:00
  • d0f9535ee7 [CI] Refactor check-bypass logic in run_tests_with_coverage (#4655) YuBaoku 2025-11-07 10:47:27 +08:00
  • 71bbedaf50 [Cherry-Pick][BugFix][CI] fix vl moe(#4867) (#4869) YuBaoku 2025-11-07 00:03:36 +08:00
  • fa28745f19 [CI] Update ERNIE-4.5-VL baseline to adapt to MoE changes (#4867) YuBaoku 2025-11-06 22:02:10 +08:00
  • cc34487810 [Feature] support mm disable_chunked (#4803) kevin 2025-11-06 21:32:25 +08:00
  • 6b68c58e8d Revert "[Bug Fix] fix ernie4_5_vl_moe (#4843)" (#4863) Jiang-Jia-Jun 2025-11-06 19:18:29 +08:00
  • 6460d4df27 [Bug Fix] fix ernie4_5_vl_moe (#4843) LokeZhou 2025-11-06 19:16:33 +08:00
  • a139f8f3cb [CI] Optimize port cleanup logic (#4860) YuBaoku 2025-11-06 19:13:48 +08:00
  • 5aa73d32f4 Update deploy.py (#4850) Zhang Yulong 2025-11-06 19:09:28 +08:00
  • 819b2dbbae Revert "【New Feature】W4afp8 supports per group quantization (#4272)" (#4854) YuBaoku 2025-11-06 17:48:28 +08:00
  • 3478d20262 [CI] Add Check PR Template (#4481) YuBaoku 2025-11-06 17:41:14 +08:00
  • b54eb7ad81 [XPU] ep+tp all2all (#4836) zhupengyang 2025-11-06 17:26:14 +08:00
  • 901d559aa7 Update README_CN.md Jiang-Jia-Jun 2025-11-06 17:19:22 +08:00
  • 0010420c56 Update README_EN.md Jiang-Jia-Jun 2025-11-06 17:19:07 +08:00
  • 83532e1d01 [Benchmark] Enhance benchmark output logging (#4682) Zhang Yulong 2025-11-06 16:53:31 +08:00
  • 095dada092 Add gemini for code review Jiang-Jia-Jun 2025-11-06 16:42:32 +08:00
  • c18b177f21 fix the get_act_fn,_load_st_projector (#4824) Echo-Nie 2025-11-06 16:13:35 +08:00
  • e4f1267186 bug: fix list to List (#4818) Echo-Nie 2025-11-06 16:13:12 +08:00
  • 89934edc10 update (#4851) Ayakouji 2025-11-06 16:08:04 +08:00
  • 6c316286c1 fix: correct typo in nvidia_gpu.md (#4848) Ding 2025-11-06 16:03:02 +08:00
  • cbe27ad9fb [Cherry-Pick] Fix ernie_vl_reasoning_parsers.py 'end_token' to 'think_end_token' (#4805) (#4842) kxz2002 2025-11-06 15:54:48 +08:00
  • 8e48da8027 [Bug Fix] process transparent image (#4807) (#4847) ApplEOFDiscord 2025-11-06 15:43:44 +08:00
  • 08ca0f6aea [Feature] [PD] add simple router and refine splitwise deployment (#4709) Juncai 2025-11-06 14:56:02 +08:00
  • 831266da7a [Fix] fix ernie4_5_vl model torch format loadding (#4447) Ayakouji 2025-11-06 14:34:21 +08:00
  • 93aedaf23e temporary fix bug of 0 size tensor (#4844) RAM 2025-11-06 14:20:59 +08:00
  • b0d213f750 fix token_type_ids for eb45-vl (#4775) RevL147 2025-11-06 14:19:57 +08:00
  • f1ea3830aa [CI] remove ernie-4_5-vl test_consistency_between_runs (#4846) YuBaoku 2025-11-06 14:19:04 +08:00
  • fc8bef2c95 [XPU][CI]Change ci vl model to 28 b (#4764) plusNew001 2025-11-06 14:12:23 +08:00
  • 354ddc8bc5 [CI] Add unittest for activation, native_paddle_backend, w4a8, w4afp8, platforms/utils (#4812) Echo-Nie 2025-11-06 14:08:00 +08:00
  • 0d7f29841d [Cherry-pick] Fix bug in develop for eb5 (#4845) chenjian 2025-11-06 14:01:42 +08:00
  • eac823aec8 [BugFix] fix total_block_num init error in worker_process (#4553) (#4817) RichardWooSJTU 2025-11-06 13:59:57 +08:00