sglang_discussions

"Skip to content\nNavigation Menu\nPlatform\nSolutions\nResources\nOpen Source\nEnterprise\nPricing\nSign in\nSign up\nsgl-project\n/\nsglang\nPublic\nNotifications\nFork 7.2k\n Star 30.3k\nCode\nIssues\n706\nPull requests\n3.6k\nDiscussions\nActions\nSecurity and quality\nInsights\nsgl-project sglang Discussions\nPinned Discussions\nSGLang Public Community Events\n📣 Announcements\n·\nsglang-bot\nSearch all discussions\nis:open \n \nClear\nSort by: Latest activity\nLabel\nFilter: Open\nCategories, most helpful, and community links\nCategories\nView all discussions\n📣\nAnnouncements\n💬\nGeneral\n💡\nIdeas\n🗳️\nPolls\n🙏\nQ&A\n🙌\nShow and tell\nMost helpful\nLast 30 days\nGautamKumarOffical\n1\nhhhFuture\n1\nCommunity links\nCode of conduct\nsglang.io\nDiscussions\n1\n🙏\nA question about Qwen-3-vl-reranker for the difference between sglang inference and transformer inference.\nSerendipity-00 asked  in Q&A · Unanswered\n 0\n1\n🙌\nE8 lattice KV cache: calibration-free 2-bit that preserves retrieval\njagmarques started  in Show and tell\n 0\n0\n💬\nUnknown diffusion LLM: DiffusionGemmaForBlockDiffusion\ngiovan001 started  in General\n 0\n1\n🙏\nHow to prioritize decode batches over prefill in SGLang? (GLM-4.7 deployment)\ncoach00 asked  in Q&A · Unanswered\n 3\n1\n🙏\nIs there a example about deepseek-v4-pro pd disaggregation？\ntaishan1994 asked  in Q&A · Unanswered\n 4\n1\n🙏\n[Diffusion] Is there support for /metrics endpoint in SGLang Diffusion (Qwen-Image)\ncapri-xiyue asked  in Q&A · Unanswered\n 1\n1\n🙏\nQuestion about serving Qwen3.5 text-only SFT model saved as Qwen3_5ForCausalLM\nnchennnn asked  in Q&A · Answered\n 2\n0\n💡\nAddition of a not-strictly-block-diffusion model\nsettheworldonfireiii started  in Ideas\n 0\n1\n🙏\nSmall commercial app use of Boson v.3\nlinguaphile asked  in Q&A · Unanswered\n 0\n2\n🙌\nPeerCache — a decentralized P2P RDMA L3 backend for SGLang HiCache\nflymysql started  in Show and tell\n 0\n1\n🙏\nDoes sglang really support run Qwen3.5-397B-A17B for processing Ultra-Long Texts(1M) ?\ndingfen asked  in Q&A · Unanswered\n 3\n📣\nSGLang Public Community Events\nsglang-bot announced  in Announcements\n 0\n2\n🙏\nDo Hopper support Deepseek V4 Flash run EP by deepep in the future?\ndingfen asked  in Q&A · Unanswered\n 0\n1\n🙏\ndeepep v2 support?\nlitiantian00 asked  in Q&A · Unanswered\n 0\n1\n💬\n[Question][Diffusion][SD3.5] Question about SGLang-native support for remaining Diffusers modules for stable-diffusion-3\nYuxingwang-intel started  in General\n 0\n1\n💡\n[RFC] IOChain: request/response filters for OpenAI-compatible serving\nmukeshbaphna started  in Ideas\n 0\n1\n💡\nمرحبا\nzakariapirate7-afk started  in Ideas\n 4\n1\n💬\nCANN 9.0.0 support?\nfiskrt started  in General\n 2\n1\n🙏\nLing 2.6 lightning/linear attention cache size not exposed in SGLang metrics/logs\nAlpaczyk asked  in Q&A · Unanswered\n 0\n1\n💡\nSupport DDTree (Diffusion Draft Tree) to further accelerate DFlash\nhuxiang088 started  in Ideas\n 2\n1\n🙏\nWhy similar CPU memory leak fixes were split across multiple PRs, and how you pinpoint leaks down to specific lines?\nShenYuhan asked  in Q&A · Unanswered\n 4\n1\n🙏\nDflash working launch parameters\nGerLinuxEnthusiast asked  in Q&A · Unanswered\n 2\n1\n🙏\nsglang在qwen3-vl-4B模型使用dflash加速但是没有收益，都是负收益\ncuihangbin asked  in Q&A · Unanswered\n 0\n1\n🙏\n[Help] sglang Docker hangs with 100% CPU & GPU Util on dual Blackwell GPUs. What's the best way to install?\nrain2307 asked  in Q&A · Unanswered\n 1\n1\n🙏\nSGLang Inference 8*H200(1 HGX). QWEN-3.5-397B-A17B-FP8\ndanunited999-del asked  in Q&A · Unanswered\n 2\nPrevious 1 2 3 4 5 … 9 10 Next\nFooter\n© 2026 GitHub, Inc.\nFooter navigation\nTerms\nPrivacy\nSecurity\nStatus\nCommunity\nDocs\nContact\nManage cookies\nDo not share my personal information"