[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

@lmsysorg Avatar @lmsysorg LMSYS Org

LMSYS Org posts on X about discussions, gpu, moe, breakthrough the most. They currently have XXXXX followers and XX posts still getting attention that total XXXXXXX engagements in the last XX hours.

Engagements: XXXXXXX #

Engagements Line Chart

Mentions: X #

Mentions Line Chart

Followers: XXXXX #

Followers Line Chart

CreatorRank: XXXXXX #

CreatorRank Line Chart

Social Influence #

Social category influence stocks #1129 technology brands #5015

Social topic influence discussions, gpu, moe, breakthrough

Top Social Posts #

Top posts by engagements in the last XX hours

"@nvidia @SemiAnalysis_ 🔗 Full Blog here:"
X Link @lmsysorg 2025-10-14T22:22Z 8501 followers, XXX engagements

"🚀 Excited to collaborate with @nvidia and @SemiAnalysis_ on pushing inference performance to the next level On the Blackwell GB200 NVL72 SGLang achieved 26K input / 13K output tokens per GPU/sec. On the @SemiAnalysis_ InferenceMAX benchmark SGLang is the default engine for DeepSeek on both NV and AMD platforms. Together we see up to X generation speedup from Hopper to Blackwell. These results reflect months of joint system-level optimization across hardware and software including PrefillDecode Disaggregation Large-Scale Expert Parallelism (EP) FP8 Attention NVFP4 GEMM kernels and"
X Link @lmsysorg 2025-10-14T22:21Z 9628 followers, 30.7K engagements

"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. ✅ Reproducible outputs across batching ✅ RL-friendly deterministic rollouts ✅ Minimal perf overhead Determinism is crucial for reproducible research debugging and true on-policy RL. SGLang makes LLM inference predictable without sacrificing too much performance. Read the full blog: 👇 #LLM #DeterministicInference #SGLang #RL"
X Link @lmsysorg 2025-09-22T21:36Z 8503 followers, 106.5K engagements

"What a night 🎉 The SGLang @lmsysorg x NVIDIA @nvidia SF Meetup brought together 400+ registered attendees (100+ waitlisted) and 500+ tuning in online. We had incredible talks and discussions on LLM inference acceleration distributed compute and open infra featuring amazing speakers from across the AI infra community. Huge thanks to @NVIDIAAI @NVIDIAAIDev and all who joined us its inspiring to see so much energy around open-source inference 💪 🎥 Missed it live Watch the full replay here: #SGLang #AIInfra #LLM #NVIDIA"
X Link @lmsysorg 2025-10-06T23:16Z 8501 followers, 12.4K engagements

"🚀 SGLang In-Depth Review of the NVIDIA DGX Spark is LIVE Thanks to @NVIDIAs early access program SGLang makes its first ever appearance in a consumer product the brand-new DGX Spark. The DGX Sparks 128GB Unified Memory and Blackwell architecture set a new standard for local AI prototyping and edge computing. We're thrilled to bring these cutting-edge performance insights and software support to the developer community. Our review dives into how to efficiently deploy and accelerate large models like Llama XXX 70B GPT-OSS using SGLang's EAGLE3 speculative decoding and @Ollama on this beautiful"
X Link @lmsysorg 2025-10-14T00:16Z 9629 followers, 311.7K engagements

"🎥 Our video review: 👉 Blog here:"
X Link @lmsysorg 2025-10-14T00:16Z 9628 followers, 8503 engagements

"@ollama's review of DGX Spark is here"
X Link @lmsysorg 2025-10-14T02:02Z 8484 followers, XXX engagements

"🚀 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 - thats a XXX / XXX speedup vs H100 settings. See the details in the 🧵 (1/4)"
X Link @lmsysorg 2025-09-25T23:11Z 8500 followers, 67K engagements