zhyncs Profile
zhyncs

@zhyncs42

Followers
758
Following
413
Statuses
301

OPINIONS ARE MY OWN, LLM Inference Acceleration, Homepage https://t.co/saCowtppUm, Software Engineer @basetenco, Member @lmsysorg SGLang, Prev @meituan @Baidu_Inc

Joined February 2024
Don't wanna be here? Send us removal request.
@zhyncs42
zhyncs
@zhyncs42
1 month
We're excited to announce SGLang @lmsysorg v0.4.1, which now supports DeepSeek @deepseek_ai V3 - currently the strongest open-source LLM, even surpassing GPT-4o.
7
15
112
@zhyncs42
zhyncs
@zhyncs42
7 hours
RT @AnushElangovan: Deepseek v3/r1 is now 3X faster than when it was released on MI300X. docker pull lmsysorg/sglang:v0.4.2.post3-rocm630 h…
0
26
0
@zhyncs42
zhyncs
@zhyncs42
15 hours
RT @__tinygrad__: .@github Actions is one of the best products in the world. CI is so important for making high quality software. https://t…
0
35
0
@zhyncs42
zhyncs
@zhyncs42
2 days
RT @hyperbolic_labs: Thanks for featuring Hyperbolic—we’re proud to partner with LMSYS and leverage SGLang for efficient large-scale model…
0
26
0
@zhyncs42
zhyncs
@zhyncs42
2 days
RT @lmarena_ai: Serving DeepSeek R1/V3 with SGLang!
0
1
0
@zhyncs42
zhyncs
@zhyncs42
2 days
RT @lmsysorg: thanks so much for adoption and better collaboration! checkout SGLang repo SGLang document for Dee…
0
5
0
@zhyncs42
zhyncs
@zhyncs42
2 days
Wow Honored to be adopted by NVIDIA!
@NVIDIAAIDev
NVIDIA AI Developer
@NVIDIAAIDev
2 days
⚡ 3.2x higher throughput in just two weeks for even more cost-effective DeepSeek-R1 #inference. 👀 Check out DeepSeek-R1 performance gains on NVIDIA Hopper GPUs, which feature powerful FP8 Tensor Cores and high-bandwidth NVLink and NVSwitch interconnects, running SGLang by @lmsysorg.
1
5
58
@zhyncs42
zhyncs
@zhyncs42
3 days
@lmsysorg @AMD @nvidia @Azure @basetenco @novita_ai_labs @BytedanceTalk @DataCrunch_io @hyperbolic_labs @Vultr @runPod Feel free to use and star the SGLang project.
1
2
12
@zhyncs42
zhyncs
@zhyncs42
3 days
@isidentical
0
0
0
@zhyncs42
zhyncs
@zhyncs42
3 days
RT @lmsysorg: The SGLang Team is honored to announce that the following well-known companies and teams, among others, have adopted SGLang f…
0
18
0
@zhyncs42
zhyncs
@zhyncs42
3 days
Proudly powered by SGLang @lmsysorg
@EmadBarsoumPi
Emad Barsoum
@EmadBarsoumPi
4 days
Running R1 on a single MI300X server!!! #AMD #ML #AI #LLM #Reasoning #Thinking #AGI
0
1
3
@zhyncs42
zhyncs
@zhyncs42
3 days
Microsoft Azure @Azure blog “Running DeepSeek-R1 on a single NDv5 MI300X VM” It’s powered by SGLang! @lmsysorg
0
1
11
@zhyncs42
zhyncs
@zhyncs42
8 days
@rogerw0108 @hsu_byron @glowsai Great work!
0
0
2
@zhyncs42
zhyncs
@zhyncs42
8 days
RT @MaitrixOrg: 🚀Unlock efficient inference-time scaling with LLM Reasoners - now powered by SGLang! > 🔥100x acceleration over previous…
0
20
0
@zhyncs42
zhyncs
@zhyncs42
9 days
@corbtt @lmsysorg
0
0
0
@zhyncs42
zhyncs
@zhyncs42
9 days
@corbtt @lmsysorg Currently, one of the providers that ranks relatively high on OpenRouter and supports DeepSeek V3 with SGLang. They have conducted performance comparisons other than DeepSeek V3, including Llama's, and SGLang still maintains a leading advantage.
1
0
3
@zhyncs42
zhyncs
@zhyncs42
9 days
@corbtt @lmsysorg yeah for sure
0
0
0
@zhyncs42
zhyncs
@zhyncs42
9 days
RT @JundeMorsenWu: Dario, the CEO of Anthropic, has now blamed all his failures on the lack of chip export controls on China after seeing D…
0
451
0
@zhyncs42
zhyncs
@zhyncs42
10 days
@michalwols @deepseek_ai @OpenAI @AnthropicAI This is great news for regular users and consumers🤣
0
0
6