LMSYS Blog · 实时热榜
- 01Win on TCO: How AMD Instinct™ MI355X Achieves Cost-Competitive Distributed Inference Through SGLang with MoRI - LMSYS Org
Win on TCO: How AMD Instinct™ MI355X Achieves Cost-Competitive Distributed Inference Through SGLang with MoRI LMSYS Org
最高第 24 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 02Highlights of SGLang at NVIDIA GTC 2026 - LMSYS Org
Highlights of SGLang at NVIDIA GTC 2026 LMSYS Org
最高第 28 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 03DeepSeek-V4 on Day 0: From Fast Inference to Verified RL with SGLang and Miles - LMSYS Org
DeepSeek-V4 on Day 0: From Fast Inference to Verified RL with SGLang and Miles LMSYS Org
最高第 26 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 04Updating 1T parameters in seconds — P2P weight transfer in Large Scale Distributed RL - LMSYS Org
Updating 1T parameters in seconds — P2P weight transfer in Large Scale Distributed RL LMSYS Org
最高第 25 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 05Heterogeneous CPU + GPU EPD Disaggregation to Boost VLM Serving - LMSYS Org
Heterogeneous CPU + GPU EPD Disaggregation to Boost VLM Serving LMSYS Org
最高第 23 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 06Improving DeepEP MoE Load Balance in SGLang with Waterfill and LPLB - LMSYS Org
Improving DeepEP MoE Load Balance in SGLang with Waterfill and LPLB LMSYS Org
最高第 17 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 07Towards Blackwell-Native 8-bit and 4-bit RL: End-to-End MXFP8 and NVFP4 RL in Miles - LMSYS Org
Towards Blackwell-Native 8-bit and 4-bit RL: End-to-End MXFP8 and NVFP4 RL in Miles LMSYS Org
最高第 9 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 08SpecForge v0.3.0: a Unified Disaggregated and Colocated Speculative Decoding Stack, and New Open SpecBundle Draft Models - LMSYS Org
SpecForge v0.3.0: a Unified Disaggregated and Colocated Speculative Decoding Stack, and New Open SpecBundle Draft Models LMSYS Org
最高第 7 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 09DSpark in SGLang: Speculative Decoding with Confidence-Driven, Variable-Length Verification - LMSYS Org
DSpark in SGLang: Speculative Decoding with Confidence-Driven, Variable-Length Verification LMSYS Org
最高第 15 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 10HPC-Ops × SGLang: High-Performance Attention, Router GEMM, and MoE Kernels from Tencent Hunyuan - LMSYS Org
HPC-Ops × SGLang: High-Performance Attention, Router GEMM, and MoE Kernels from Tencent Hunyuan LMSYS Org
最高第 5 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 11Accelerating SGLang HiCache with Netpreme X-Mem™ MPU - LMSYS Org
Accelerating SGLang HiCache with Netpreme X-Mem™ MPU LMSYS Org
最高第 14 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 12ROCm Support for Miles: Large-Scale RL Post-Training on AMD Instinct™ GPUs - LMSYS Org
ROCm Support for Miles: Large-Scale RL Post-Training on AMD Instinct™ GPUs LMSYS Org
最高第 29 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 13Serving GLM5.2 NVFP4 Agentic Workload with SGLang: Reaching 500 TPS in 2 Weeks - LMSYS Org
Serving GLM5.2 NVFP4 Agentic Workload with SGLang: Reaching 500 TPS in 2 Weeks LMSYS Org
最高第 13 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 14RadixArk Joins Forces with Google to Bring Full SGLang Features to TPUs - LMSYS Org
RadixArk Joins Forces with Google to Bring Full SGLang Features to TPUs LMSYS Org
最高第 8 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 15HiSparse: Turbocharging Sparse Attention with Hierarchical Memory - LMSYS Org
HiSparse: Turbocharging Sparse Attention with Hierarchical Memory LMSYS Org
最高第 27 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 16SGLang Adds Day-0 Support for Muse Glimmer, a Multimodal Model Built for Local Agentic Workflows - LMSYS Org
SGLang Adds Day-0 Support for Muse Glimmer, a Multimodal Model Built for Local Agentic Workflows LMSYS Org
最高第 4 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 17Unlocking 25x Inference Performance with SGLang on NVIDIA GB300 NVL72 - LMSYS Org
Unlocking 25x Inference Performance with SGLang on NVIDIA GB300 NVL72 LMSYS Org
最高第 30 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 18SGLang and Miles Add Day-0 Support for Kimi K3 - LMSYS Org
SGLang and Miles Add Day-0 Support for Kimi K3 LMSYS Org
最高第 11 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 19Announcing the Recipient of the 2026 LMSYS PhD Fellowship - LMSYS Org
Announcing the Recipient of the 2026 LMSYS PhD Fellowship LMSYS Org
最高第 20 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 20Full-Stack Performance Optimization of AR+DiT in SGL-Diffusion - LMSYS Org
Full-Stack Performance Optimization of AR+DiT in SGL-Diffusion LMSYS Org
最高第 6 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 21MOSS-TTS Local Transformer v1.5 on SGLang-Omni: Serving Native-Streaming 48 kHz Speech - LMSYS Org
MOSS-TTS Local Transformer v1.5 on SGLang-Omni: Serving Native-Streaming 48 kHz Speech LMSYS Org
最高第 18 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 22Higgs Audio v3 TTS on SGLang-Omni: Real-Time, Controllable Speech for Voice Agents - LMSYS Org
Higgs Audio v3 TTS on SGLang-Omni: Real-Time, Controllable Speech for Voice Agents LMSYS Org
最高第 21 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 23SGLang and Miles Add Day-0 Support for Qwen3.8 - LMSYS Org
SGLang and Miles Add Day-0 Support for Qwen3.8 LMSYS Org
最高第 1 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 24OPD Support in Miles - LMSYS Org
OPD Support in Miles LMSYS Org
最高第 12 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 25Unified Radix Cache: One Tree for Hybrid Model Prefix Caching - LMSYS Org
Unified Radix Cache: One Tree for Hybrid Model Prefix Caching LMSYS Org
最高第 2 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 26SGLang Adds Day-0 Support for NVIDIA Nemotron 3.5 Lightning - LMSYS Org
SGLang Adds Day-0 Support for NVIDIA Nemotron 3.5 Lightning LMSYS Org
最高第 3 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 27Toward a Cleaner Quantization Stack in SGLang - LMSYS Org
Toward a Cleaner Quantization Stack in SGLang LMSYS Org
最高第 10 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 28SGLang and Miles Add Day-0 Support for NVIDIA Nemotron 3 Ultra for Long-Running Autonomous Agents - LMSYS Org
SGLang and Miles Add Day-0 Support for NVIDIA Nemotron 3 Ultra for Long-Running Autonomous Agents LMSYS Org
最高第 22 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 29Agent-Assisted SGLang Development: An Initial Exploration - LMSYS Org
Agent-Assisted SGLang Development: An Initial Exploration LMSYS Org
最高第 16 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分 - 30The next generation of speculative decoding: DFlash and Spec V2 - LMSYS Org
The next generation of speculative decoding: DFlash and Spec V2 LMSYS Org
最高第 19 名00:00 达到当日首次采集时已在榜当日结束时仍在榜累计约16小时17分