Junchen Jiang
@JunchenJiang
Followers
397
Following
88
Media
4
Statuses
85
CS Prof @ UChicago https://t.co/U01oOWGnip (Fast distributed LLM inference) https://t.co/hoetjwXKIt (Best KV cache layer)
Chicago, IL
Joined September 2012
Go LMCache ๐.
1
1
10
RT @lmcache: 8 KV-Cache Systems You Canโt Afford to Miss in 2025. By 2025, KV-cache has evolved from a โnice-to-haveโ optimization into a cโฆ.
0
16
0
RT @TerryTangYuan: Excited to share that I'll be speaking at ๐๐น๐ผ๐๐ฑ ๐ก๐ฎ๐๐ถ๐๐ฒ ๐๐ด๐ ๐๐ ๐๐ฎ๐, in addition to @KubeCon_! . Dan Sun and I will be delโฆ.
colocatedeventsna2025.sched.com
View more about this event at CNCF-hosted Co-located Events North America 2025
0
4
0
RT @lmcache: CacheGen( lets you store KV caches on disk or AWS S3 and load them way faster than recomputing! . Modeโฆ.
0
6
0
RT @bentomlai: ๐คย What is KV cache offloading and why does it matter for LLM inference?. #LLMs use the KV cache to accelerate inference speeโฆ.
0
2
0
RT @lmcache: LMCache supports gpt-oss (20B/120B) on Day 1!. TTFT 1.20s โ 0.39s (-67.5%), finish time 15.70s โ 7.73s (-50.7%) compared to Vaโฆ.
0
9
0
RT @lmcache: ๐ Big news from LMCache Lab!. ๐ 3 papers accepted at SOSP โ25 & NSDI โ26, pushing the frontier of LLM-inference efficiency:โฆ.
0
6
0
RT @NadavTimor: KV cache go brrr with @JunchenJiang's @lmcache!.Join us tomorrow to learn more about nextโgen longโcontext LLM inference: hโฆ.
0
2
0
RT @zhzHNN: @hidecloud Hi, we are organizing a meetup in Bay Area to discuss context engineering with @JunchenJiang and @lmcache . Are youโฆ.
0
1
0
RT @siddhantrayyy: With RAG and agents becoming ubiquitous in LLM systems, tuning quality and performance JOINTLY is essential to achieve tโฆ.
0
6
0
RT @this_will_echo: ๐คฏ Believe it or not, even when an LLM generates just ONE SINGLE word, it can still be powerful!. Say in recommendation:โฆ.
0
7
0
RT @lmcache: ๐จ LMCache now turbocharges multimodal models in vLLM!. By caching image-token KV pairs, repeated images now get ~100% cache hiโฆ.
0
12
0
RT @lmcache: ๐๐ ๐๐ฎ๐ฐ๐ต๐ฒ ๐ฟ๐ฒ๐ฎ๐ฐ๐ต๐ฒ๐ ๐ฎ,๐ฌ๐ฌ๐ฌ+ ๐๐๐ฎ๐ฟ๐ ๐ผ๐ป ๐๐ถ๐๐๐๐ฏ! ๐ . A huge thank you to our open-source communityโyour support is fueling nextโgen effโฆ.
0
5
0
RT @GitHubGPT: ๐ LMCache.๐ง LMCache, an LLM engine, boosts performance by minimizing TTFT and enhancing throughput via effective KV cache maโฆ.
github.com
Supercharge Your LLM with the Fastest KV Cache Layer - LMCache/LMCache
0
3
0
RT @lmcache: Our very own @JunchenJiang gave a talk about large-scale efficient inference at Open Source Summit 2025 yesterday with Yue Zhโฆ.
0
2
0