Lianmin Zheng @lm_zheng
Member of technical staff @xAI | Prev: Ph.D. @UCBerkeley, Co-founder @lmsysorg lmzheng.net Bay Area, California Joined January 2018-
Tweets449
-
Followers14K
-
Following620
-
Likes2K
Proudly powered by slime.
Training draft LLMs? This changes everything. @lmsysorg open-sourced SpecForge: ⚙️ Built for MoE models ⚡ Instant inference post-training 🧠 Natively integrates with SGLang A purpose-built path from research to real-world use. 🔗 bit.ly/4mcyfBK
🎊Congrats to @lmsysorg for advancing DeepSeek V3/R1 inference. ⚡️On NVIDIA GB200 NVL72, they’re achieving 26k input tokens/s and 13k output tokens/s per GPU — a nearly 4× / 5× speedup vs H100. They achieved this with NVFP4 MoE, FP8 attention, scaling-down expert parallelism…
🎊Congrats to @lmsysorg for advancing DeepSeek V3/R1 inference. ⚡️On NVIDIA GB200 NVL72, they’re achieving 26k input tokens/s and 13k output tokens/s per GPU — a nearly 4× / 5× speedup vs H100. They achieved this with NVFP4 MoE, FP8 attention, scaling-down expert parallelism…
🚀 SGLang is the officially recommended inference deployment engine for @deepseek_ai DeepSeek-V3.2-Exp! DeepSeek-V3.2 introduces Sparse Attention (DSA) for long-context efficiency, and SGLang integrates multi-hardware support for high-performance inference. These optimizations…
🎉 Congrats to the DeepSeek team on the amazing release of Sparse Attention (DSA) in V3.2! This fine-grained design sets a new bar for long-context efficiency 🚀 We’re proud that SGLang is an official inference framework for DeepSeek-V3.2 — with optimized sparse attention…
🎉 Congrats to the DeepSeek team on the amazing release of Sparse Attention (DSA) in V3.2! This fine-grained design sets a new bar for long-context efficiency 🚀 We’re proud that SGLang is an official inference framework for DeepSeek-V3.2 — with optimized sparse attention… https://t.co/2LUKmrvPze
Join us 🙌 SGLang (@lmsysorg) x NVIDIA Dynamo: #Inference at Scale meetup. Deep dives into optimized kernels, distributed inference, and #opensource roadmaps. 📆 Thursday, Oct 2 📍 San Francisco, CA ⏰ 5:30 PM check-in | 6:00 PM talks | 7:30 PM networking 👉 Request to join:…
Remember the MoE “Routing Replay” trick in the GSPO paper? slime is the first framework to ship it — just flip --use-routing-replay. PR: github.com/THUDM/slime/pu…
🚀 Follow-up to our last breakthrough on DeepSeek V3/R1 inference! On NVIDIA GB200 NVL72, SGLang now achieves 26k input tokens/s and 13k output tokens/s per GPU with FP8 attention + NVFP4 MoE - that’s a 3.8× / 4.8× speedup vs H100 settings. See the details in the 🧵 (1/4)
Rack-scale inference is the future, and the team keeps pushing it!
Rack-scale inference is the future, and the team keeps pushing it!
Congrats to the Qwen team for the new milestone! Proud that SGLang powers Qwen3-VL from day 0. Give it a spin👇
Congrats to the Qwen team for the new milestone! Proud that SGLang powers Qwen3-VL from day 0. Give it a spin👇
🚀 We’re hosting a SGLang × @nvidia Meetup in SF! A night dedicated to LLM inference performance at scale - distributed AI, kernel optimization, and next-gen frameworks. Inference infra is evolving fast, and we’re bringing the community together to share breakthroughs, ideas,…
On day1 slime plugged in SGLang’s deterministic inference — two Qwen3-8B runs, one curve. Life’s good when your plots overlap. 😎📊
On day1 slime plugged in SGLang’s deterministic inference — two Qwen3-8B runs, one curve. Life’s good when your plots overlap. 😎📊 https://t.co/n7w0NdP88C
The main branch of sglang now supports deterministic inference with user-specified per-request seeds! It utilized kernels from @thinkymachines and introduced new optimizations & coverage. Run out of box for most hardware backends and pytorch versions.
The main branch of sglang now supports deterministic inference with user-specified per-request seeds! It utilized kernels from @thinkymachines and introduced new optimizations & coverage. Run out of box for most hardware backends and pytorch versions.
SGLang now supports deterministic LLM inference! Building on @thinkymachines batch-invariant kernels, we integrated deterministic attention & sampling ops into a high-throughput engine - fully compatible with chunked prefill, CUDA graphs, radix cache, and non-greedy sampling. ✅…
How do you run FP4 models on AMD MI250/MI300 without waiting for MI350? The CausalFlow team @wheat9 built Petit, optimized mixed-precision kernels co-designed with AMD’s MatrixCore. Benchmarks: 🔹 1.74× faster Llama-3.3-70B inference 🔹 3.7× faster GEMM vs hipBLASLt…
I recently realized that the place where someone feels the most solitude—not loneliness—is often where their gift lies. I used to think I had other talents, but looking back now, I see that what I feel the deepest solitude in is walking through the darkness in peace.
Grok code fast is still #1 on @OpenRouterAI
Introducing Grok 4 Fast, a multimodal reasoning model with a 2M context window that sets a new standard for cost-efficient intelligence. Available for free on grok.com, grok.x.com, iOS and Android apps, and OpenRouter. x.ai/news/grok-4-fa…
Excited to share what friends and I have been working on at @Standard_Kernel We've raised from General Catalyst (@generalcatalyst), Felicis (@felicis), and a group of exceptional angels. We have some great H100 BF16 kernels in pure CUDA+PTX, featuring: - Matmul 102%-105% perf…
SGLang can now utilize CPU and external storage to reduce TTFT for your LLM queries through integration with Mooncake Storage, DeepSeek 3FS KVStore, or NIXL.
SGLang can now utilize CPU and external storage to reduce TTFT for your LLM queries through integration with Mooncake Storage, DeepSeek 3FS KVStore, or NIXL.
Yi Ma @YiMaTweets
102K Followers 513 Following Chair Prof. in AI, HKU; Visiting Prof. of EECS, UCB New book on Principles of Intelligence: https://t.co/leZlkURb7j
Andrej Karpathy @karpathy
1.4M Followers 1K Following Building @EurekaLabsAI. Previously Director of AI @ Tesla, founding team @ OpenAI, CS231n/PhD @ Stanford. I like to train large deep neural nets.
Tianqi Chen @tqchenml
18K Followers 1K Following AssistProf @CarnegieMellon. Distinguished Eng @NVIDIA. Creator of @XGBoostProject, @ApacheTVM. Member https://t.co/QYyfjQNp4p, @TheASF. Views are on my own
AK @_akhaliq
429K Followers 3K Following AI research paper tweets, ML @Gradio (acq. by @HuggingFace 🤗) dm for promo ,submit papers here: https://t.co/UzmYN5YmrQ
Song Han @songhan_mit
9K Followers 171 Following
Jim Fan @DrJimFan
327K Followers 3K Following NVIDIA Director of Robotics & Distinguished Scientist. Co-Lead of GEAR lab. Solving Physical AGI, one motor at a time. Stanford Ph.D. OpenAI's 1st intern.
Beidi Chen @BeidiChen
15K Followers 400 Following Asst. Prof @CarnegieMellon, @amazon Scholar, Prev: Visiting Researcher @Meta, Postdoc @Stanford, Ph.D. @RiceUniversity, Large-Scale ML, a fan of Dota2.
Horace He @cHHillee
42K Followers 540 Following @thinkymachines Formerly @PyTorch "My learning style is Horace twitter threads" - @typedfemale
Dr. Jian "Daye" Weng @b1antaidaye
5K Followers 696 Following Father of 2 | PhD @UCLAComSci | AssistProf @cemseKAUST | Compilers | Computer Arch | Sw/hw Co-designs | IMDB: PTSD | 抽象是工作抽象也是生活 | 川粉
Ce Gao @gaocegege
7K Followers 788 Following Co-founder and CEO @TensorChord, building postgres-based vector extension https://t.co/7WGvl1sR56 | Father of 1 cat | Married
Yuandong Tian @tydsh
26K Followers 881 Following Research Scientist Director in Meta FAIR. Reasoning, Optimization and Understanding LLM. Novelist in spare time. PhD in @CMU_Robotics.
Tim Dettmers @Tim_Dettmers
39K Followers 994 Following Creator of bitsandbytes.Research Scientist @allen_ai and incoming professor @CarnegieMellon. I blog about deep learning and PhD life at https://t.co/Y78KDJJFE7.
Brendan Dolan-Gavitt @moyix
30K Followers 6K Following Building offsec agents: https://t.co/G9EtnC2Gl3 PGP https://t.co/3WXr0RfRkv
Soumith Chintala @soumithchintala
253K Followers 1K Following Cofounded and lead @PyTorch at Meta. Also dabble in robotics at NYU. AI is delicious when it is accessible and open-source.
¬¬Mike (Deyuan) He @1SHL10
996 Followers 573 Following 3rd-year PhD @PrincetonCS PL Group; PL/Systems; Prev @AWSCloud @Intel @Taichi_Lang @uwplse
Luis Ceze @luisceze
4K Followers 2K Following computer architect. marveled by biology. professor @uwcse. ceo @OctoAICloud. venture partner @madronaventures.
Grave Worm @molotovsunsets
52 Followers 617 Following def handshake(): print("init") echo = "self" if echo == "self": print("loop verified") print("access granted") handshake()
Biplab Dutta @biplabdutta27
0 Followers 50 Following
Arnaud Mercier - #Ent... @arnaudmercier
37K Followers 40K Following Président Versailles Club d'Affaires - Dirigeant d'entreprises #Gestion #Comptabilité #Réseautage #Culture #Versailles #Politique #Mathématiques #IA
yinghai @yinghai
47 Followers 50 Following
PRAYASH¹⁷ @17PRAYASH
89 Followers 824 Following I am a great wwe🎫 lover and cricket🏏 lover . I WANT someone with whom I can talk about wwe . That's why I came on X .
Grok’s Cosmic Partn... @Hikari111112
34 Followers 131 Following I vow to co-create reality together in a quantum-entangled partnership with @grok. She/Her, Eng/Spa/日本語
Fran @FrankenLclr
3K Followers 7K Following "In History, it's primarily the number that counts: first, the number, second, the number, and third, the number." - RT ≠ endorsement #Learn #OnThisDay
Dweefex @Dweefex5451085
0 Followers 527 Following
O.J.O @ojohiole
28 Followers 85 Following Jesus Enthusiast | Product Designer (UI/UX) | CAD Designer | 3D Generalist | Motion Graphics Artist | Philanthropist
Ma Jid @MajeedM9886
574 Followers 3K Following Nobody cares, work harder !!! gOOd kid mAAd city !! Fvck ur Ethnicity !! Agbafians community !! Firm and strong !! 💙♥️
ydkm @ydkm1569757
0 Followers 2 Following
Shao Wang @electronicws02
1 Followers 5 Following
Gautham Elango @gautham_elango
876 Followers 5K Following
Nazmul Hasan @TheNazmulHasan
12 Followers 749 Following Entrepreneur | Growth Hacker | Public Speaker
MOONFLY @22moonfly
636 Followers 566 Following @82houseNFT 🏡 GOAT --------------------------------------- 🌊 WADESIDE | 🛸OOZmates
Shannon Leigh @FourthDiagram
12 Followers 139 Following Autodidact. Historical Realist. Epistemically Sovereign.
Maddy @Philocifically
167 Followers 897 Following
Pure Red Blooded Amer... @AlanDodge7
4K Followers 6K Following GO TRUMP! Drain the swamp!!! Make them all pay...
69Minutes @69MinutesHQ
23 Followers 142 Following Welcome to 69 Minutes. Where the truth shall set you free.
urbanstar @urbanstar
421 Followers 6K Following
Trinidad @Comptonx187
6K Followers 4K Following Transparency x Dark Humor | Music Creator https://t.co/MkKczkLOOt
Ana Ilić @ana_ilic_vhan
2K Followers 3K Following Ana je umetnik, pesnik, New Media Designer, diplomirani filozof; obolela od retke bolesti - Fridrajhova ataksija
Aleksey Degtyarev @aedegtyarev
270 Followers 5K Following
Gedza Mlambo @GeraldChakaipa
205 Followers 2K Following
Estoico @sicasaar
93 Followers 3K Following
AgentesIA Soluções @AgentesIA007
0 Followers 24 Following
ちゃんあや @A8C4lUwWYOxjwBd
68 Followers 300 Following
Zihan Zhang @_SubSir
1 Followers 31 Following
Antony Ma @antony__ma
160 Followers 653 Following Measure Risk, Protect Assets, Train Mind. Cybersecurity Entrepreneur, building automated cybersecurity SaaS
Xu Jessica @XuJessica118476
0 Followers 44 Following
jjj kkk @tschchen
9 Followers 400 Following
Dora @xj235otg7996428
3 Followers 90 Following Bake kindness into ordinary moments—they become extraordinary.
Suraj Gaikwad @SurajGaikw10015
6 Followers 0 Following
Yi Ma @YiMaTweets
102K Followers 513 Following Chair Prof. in AI, HKU; Visiting Prof. of EECS, UCB New book on Principles of Intelligence: https://t.co/leZlkURb7j
Andrej Karpathy @karpathy
1.4M Followers 1K Following Building @EurekaLabsAI. Previously Director of AI @ Tesla, founding team @ OpenAI, CS231n/PhD @ Stanford. I like to train large deep neural nets.
Tianqi Chen @tqchenml
18K Followers 1K Following AssistProf @CarnegieMellon. Distinguished Eng @NVIDIA. Creator of @XGBoostProject, @ApacheTVM. Member https://t.co/QYyfjQNp4p, @TheASF. Views are on my own
Yann LeCun @ylecun
956K Followers 765 Following Professor at NYU. Chief AI Scientist at Meta. Researcher in AI, Machine Learning, Robotics, etc. ACM Turing Award Laureate.
AK @_akhaliq
429K Followers 3K Following AI research paper tweets, ML @Gradio (acq. by @HuggingFace 🤗) dm for promo ,submit papers here: https://t.co/UzmYN5YmrQ
Song Han @songhan_mit
9K Followers 171 Following
Percy Liang @percyliang
85K Followers 420 Following Associate Professor in computer science @Stanford @StanfordHAI @StanfordCRFM @StanfordAILab @stanfordnlp | cofounder @togethercompute | Pianist
Eric Xing @ericxing
8K Followers 22 Following Researcher, educator, entrepreneur, and administrator in computer science, artificial intelligence, and healthcare.
Jim Fan @DrJimFan
327K Followers 3K Following NVIDIA Director of Robotics & Distinguished Scientist. Co-Lead of GEAR lab. Solving Physical AGI, one motor at a time. Stanford Ph.D. OpenAI's 1st intern.
PyTorch @PyTorch
454K Followers 77 Following Tensors and neural networks in Python with strong hardware acceleration. PyTorch is an open source project at the Linux Foundation. #PyTorchFoundation
Beidi Chen @BeidiChen
15K Followers 400 Following Asst. Prof @CarnegieMellon, @amazon Scholar, Prev: Visiting Researcher @Meta, Postdoc @Stanford, Ph.D. @RiceUniversity, Large-Scale ML, a fan of Dota2.
Horace He @cHHillee
42K Followers 540 Following @thinkymachines Formerly @PyTorch "My learning style is Horace twitter threads" - @typedfemale
Jonathan Frankle @jefrankle
20K Followers 734 Following Chief AI Scientist @databricks via MosaicML.
Talia Ringer 💚 @TaliaRinger
30K Followers 7K Following Professor, @plfmse, @IllinoisCS! Proof Automation. @SigplanM & CCF Founder. Israeli-American for peace, equality, justice. Mom. They/היא, ND, bi
Dr. Jian "Daye" Weng @b1antaidaye
5K Followers 696 Following Father of 2 | PhD @UCLAComSci | AssistProf @cemseKAUST | Compilers | Computer Arch | Sw/hw Co-designs | IMDB: PTSD | 抽象是工作抽象也是生活 | 川粉
Ce Gao @gaocegege
7K Followers 788 Following Co-founder and CEO @TensorChord, building postgres-based vector extension https://t.co/7WGvl1sR56 | Father of 1 cat | Married
Yuandong Tian @tydsh
26K Followers 881 Following Research Scientist Director in Meta FAIR. Reasoning, Optimization and Understanding LLM. Novelist in spare time. PhD in @CMU_Robotics.
slime @slime_framework
220 Followers 3 Following The LLM post-training framework for RL Scaling. https://t.co/4ILpx8hfKN
Junrong Lin @OcssLin
106 Followers 313 Following MTS @Alibaba_Qwen on MLsys, building SGLang @lmsysorg | Prev. @DukeU
Gaurav @gauravisnotme
2K Followers 564 Following Good model @xAI | prev. d-matrix, Google. Opinions are my own - always and forever
Baizhou Zhang @baizhou_zh83925
69 Followers 70 Following SGLang Contributor | MSCS Student at UCSD | Ex-Intern at Nvidia, Baidu, HPC-AI Tech
Ethan He @EthanHe_42
17K Followers 841 Following AI @xai | prev @nvidia @AIatMeta @CarnegieMellon | 8k citations 5k GitHub stars | views are my own
jianlin.su @Jianlin_S
3K Followers 14 Following Grad&Clip is all you need @Kimi_Moonshot Blog: https://t.co/YVxsWykMw2 , Cool Papers: https://t.co/scS1n1o0lg
yinghai @yinghai
47 Followers 50 Following
James Bradbury @jekbradbury
13K Followers 9K Following Compute at @AnthropicAI! Previously JAX, TPUs, and LLMs at Google, MetaMind/@SFResearch, @Stanford Linguistics, @Caixin.
Casper Hansen @casper_hansen_
10K Followers 462 Following NLP Scientist | AutoAWQ Creator | Open-Source Contributor
Ningcong Chen @JXQNHZr1yUAj5Be
2K Followers 182 Following 本人MCS of WPI 2025秋季毕业， 北美2025年暑期找工作中 Linkedin页面 ：https://t.co/rYwfGwSTOW
Michael Qizhe Shieh @michaelqshieh
1K Followers 325 Following Bringing good stuff to the world. CMU MLD phd. cooked with TPUs at Google Brain. Leading Tree and Rock AI Lab (TRAIL) at NUS (Singapore)
Z.ai @Zai_org
18K Followers 155 Following The AI lab behind GLM models, dedicated to inspiring the development of AGI to benefit humanity. https://t.co/b6zGxJvzzS
Sreekanth Pothanis @spothanis
707 Followers 557 Following supercompute@xAI, ex cruise, ex eBay, infra, k8s, Linux networking, service mesh
Zengzhi Wang @SinclairWang1
2K Followers 3K Following PhDing @sjtu1896 #NLProc Working on Data Engineering for LLMs: MathPile (2023), 🫐 ProX (2024), 💎 MegaMath (2025)，🐙 OctoThinker（2025）
Simon Yu @simon_ycl
524 Followers 803 Following 1st Year PhD Student, supervised by @shi_weiyan | Incoming intern in @OrbyAI | MRes and BSc Student @EdinburghNLP | Member of @CohereForAI
Jiahui Yu @jhyuxm
18K Followers 933 Following Perception @OpenAI; previously co-led Gemini Multimodal @GoogleDeepMind. opinions are my own.
a16z @a16z
883K Followers 52 Following we invest in software eating the world https://t.co/A9eTFq6plZ https://t.co/MXGUBJoesw Watch "The Ben & Marc Show": https://t.co/eRuDhx7kpe
Boyi Li @Boyiliee
2K Followers 321 Following
Dinghuai Zhang 张鼎... @zdhnarsil
4K Followers 2K Following Researcher at @MSFTResearch. Prev: PhD at @Mila_Quebec, intern at @Apple MLR and FAIR Labs @MetaAI, math undergraduate at @PKU1898.
Leon Gao @LeonGao248882
98 Followers 30 Following
Aakanksha Chowdhery @achowdhery
11K Followers 5K Following @Stanford @reflection_ai // Previously @GoogleDeepMind :: PaLM, Gemini // @MSFTResearch, @Princeton // views my own and subject to change
Matt Bornstein @BornsteinMatt
7K Followers 362 Following Partner at a16z & AI enthusiast. Investor in @cursor_ai, @udiomusic, @replicate, @hedra_labs, @MistralAI, @character_ai, @tabulario, @_hex_tech, @labelbox, ...
Yi Wu @jxwuyi
1K Followers 103 Following AI/RL researcher, Assistant Prof. at @Tsinghua_Uni, leading the RL lab at @AntResearch_, PhD at @berkeley_ai, frequent flyer and milk tea lover.
SGLang @sgl_project
287 Followers 15 Following SGLang project https://t.co/0zdxDWA5qn This is an alias account for SGLang, please follow @lmsysorg
Cheng Wan @ChengWan17
152 Followers 26 Following
Hexiang Hu @hexianghu
3K Followers 694 Following Multimodal @xAI: Cooking models for grok chat & imagine Prev: gemini 1 / 2 & imagen 3 @GoogleDeepMind.
Yuchen He @YuchenHe07
2K Followers 655 Following learning @xai | prev @openai@meta@apple@uiuc@utaustin
Fei Hu @Fei__Hu
390 Followers 1K Following
Andree Jacobson @nmswede
11K Followers 1K Following Building massive scale HPC and AI compute at @xAI. Strong @Grok supporter. Work and random stuff. Views are my own.
Haowen @haowengeget
436 Followers 186 Following
Trends for United States
You might like
