Nice example of a production #vLLM setup on 𝗡𝗲𝗯𝗶𝘂𝘀 with terraform, managed K8s, inference, and observability all in one place.

This can be a ref stack builders can use without reinventing the basics 💡.
👨🏻‍💻 full code on our repo.
github.com/CloudThrill/vllm-production-stack-terraform

kv_cache Explained: How It Enhances vLLM Inference - Cloudthrill This blog is my attempt to break it down simply, without drowning in dark math :). If you’ve ever wondered what kv_cache actually does, you’re in the right place. Let’s make it click. explore how KV cache enhances vllm inference

🏆And our #1 - 2025 blog-post on @Cloudthrill is… KV Cache explained (𝗟𝗶𝗸𝗲 𝗜’𝗺 𝟱)

Ever wondered what #KVCache really is in LLM inference?
Here's the simplest analogy for beginners plus an overview of popular KV cache optimization techniques!

📖 cloudthrill.ca/kv_cache-exp...

🏆Ranked #2 most-read in 2025 - #vLLM for Beginners (Key features)
2️⃣ Here’s the most exhaustive list of VLLM features you wish you knew. 👇
📖 cloudthrill.ca/what-is-vllm...

Learn what makes #vllm the 𝗥𝗼𝗹𝗹𝘀 𝗥𝗼𝘆𝗰𝗲 of Inference in production✨. #vLLM #AIForBeginners

🏆Ranked #3: 𝗟𝗟𝗠 𝗤𝘂𝗮𝗻𝘁𝗶𝘇𝗮𝘁𝗶𝗼𝗻, because cheaper inference is never just about INT8 vs FP16.

Here’s everything you wish you knew about LLM quantization.👇
📖 cloudthrill.ca/llm-quantization-all-you-need-to-know
🎙️Podcast (YouTube):"from 𝘎𝘎𝘜𝘍 𝘵𝘰 enterprise 𝘘𝘶𝘢𝘯𝘵ization"♥️
📺 youtube.com/watch?v=XTE0oS7b6fM

🎄This week, we’re counted down CloudThrill’s 𝗧𝗼𝗽 𝟯 𝗺𝗼𝘀𝘁-𝗿𝗲𝗮𝗱 blog posts of 𝟮𝟬𝟮𝟱🏆.
Three posts. Three lessons. First reveal drops this Monday 👀

#CloudThrill #LLM #AIInfrastructure #LLM #OpenSourceAI #AIEngineering

CloudThrill is a proud sponsor of Tech Beats Unplugged podcast🎙️. 🔥New episode out with Michael (WebScale) Webster- breaking down the VMware–Broadcom chaos, Nutanix , and real exit strategies. Listen now🎧👇🏼.

This terraform stack delivers a production-ready vLLM serving environment On @awscloud.bsky.social #EKS, supporting both CPU/GPU inference with operational best practices embedded in AWS Integration and Automation (𝗮𝘄𝘀-𝗶𝗮). A One Click Deploy🔥Check the repo and Blog below 👇🏻

This terraform stack delivers a production-ready vLLM serving environment On 𝗚𝗼𝗼𝗴𝗹𝗲 𝗖𝗹𝗼𝘂𝗱 𝗚𝗞𝗘, supporting both 𝗖𝗣𝗨/𝗚𝗣𝗨 inference with operational best practices embedded in #Terraform #GKE Module. A One Click Deploy🔥Check the repo and blog below 👇🏻

🔥Check out what’s cooking in #vLLM for 𝟮𝟬𝟮𝟲 and beyond. From the project leader himself 𝗦𝗶𝗺𝗼𝗻 𝗠𝗼! #𝗢𝗽𝗲𝗻𝗦𝗼𝘂𝗿𝗰𝗲𝗔𝗜 #LeadingTheway 💪 #RaySummit2025 #Anyscale

Still thinking of hosting your on AI Backend?
our FREE vLLM POC is still live - but not forever.
📢𝗔𝗽𝗽𝗹𝘆 𝗻𝗼𝘄 → cloudthrill.ca/ai-poc

Run AI assistants, RAG, or open models privately in the cloud:
✅ No external APIs
✅ No vendor lock-in
✅ Total data control

Your Infra. Your Models. Your rules.🏆🏁

💡 In this 5-min read you'll learn:
✅ How embeddings work – in the simplest way possible
🔁 Chunk sizes, overlaps, and text splitters
📦 Vector DBs, popular embedding models used today

💡Oh,& don’t forget, our Private 𝗔𝗜 𝗜𝗻𝗳𝗲𝗿𝗲𝗻𝗰𝗲 campaign is still running, with a 𝐋𝐈𝐌𝐈𝐓𝐄𝐃 FREE 𝐏𝐎𝐂 cloudthrill.ca/ai-poc

🍁We’re excited to share that CloudThrill has been awarded a 𝐏𝐫𝐨𝐒𝐞𝐫𝐯𝐢𝐜𝐞𝐬 𝐩𝐫𝐞𝐪𝐮𝐚𝐥𝐢𝐟𝐢𝐜𝐚𝐭𝐢𝐨𝐧👏🏼 with 𝐏𝐮𝐛𝐥𝐢𝐜 𝐒𝐞𝐫𝐯𝐢𝐜𝐞𝐬 𝐚𝐧𝐝 𝐏𝐫𝐨𝐜𝐮𝐫𝐞𝐦𝐞𝐧𝐭 𝐂𝐚𝐧𝐚𝐝𝐚 !
👋🏻 Work with a 𝐩𝐮𝐛𝐥𝐢𝐜 𝐚𝐠𝐞𝐧𝐜𝐲 ? Let’s talk about your challenges - we’d love to hear from you! cloudthrill.ca/contact-us
#CloudThrill #ProServices #GovernmentOfCanada

Check out our new 𝐯𝐋𝐋𝐌 𝐏𝐫𝐨𝐝𝐮𝐜𝐭𝐢𝐨𝐧 𝐒𝐭𝐚𝐜𝐤 blog👇🏼
💎We cover:
✅ What is 𝐯𝐋𝐋𝐌 production stack ?
✅ Request Flow & Architecture breakdown
✅ Serving Engine, Request Router & KV-Cache Netwrk
✅ Autoscaling & built-in fault-tolerance
✅ One-click Helm install

#LLMs #Kubernetes #Cloudthrill #vLLM

Here’s a full recap from #CloudThrill team of the vLLM beginners series, broken in 3 parts 💫 share and enjoy!

🔐 Learn the key to easy and production-grade secret management on K8s 👇🏼
𝐋𝐢𝐤𝐞 𝐭𝐡𝐢𝐬 𝐤𝐢𝐧𝐝 𝐨𝐟 𝐬𝐭𝐮𝐟𝐟? Subscribe here 👉 tinyurl.com/CloudThrillBlogs

Get your teams to level up their CI/CD skills with this GithubActions cert guide 👇🏻

#NewBlog: final part of our #VLLM blog series🔥
💎This, we shift from theory to practice, covering #vLLM installs across platforms? check our new blog, where we break it down in 5 sections😎#TherYouGo

New week, new blog! 👇🏼

👋🏼See you next Thursday! #Livestream #LLMs #Quantization

#NewBlog part 2 of our #VLLM blog series 🔥
💎What makes #VLLM the Rolls Royce of inference? 👇🏻check our new blog, where we break it down in 5 performance-packed layers😎#TherYouGo

Terraform Pipelines for Dummies Part3: GitHub Actions Azure Deploy with OIDC - Cloudthrill Struggling with Azure credential management in your CI/CD pipelines? Both Azure and GitHub Actions now support OpenID Connect (OIDC) for secure deployments by simplifying the process, and aligning with modern security practices. With GitHub’s OIDC provider, you can authenticate directly from your workflows using managed identities without the need for static access keys.

🚀#NewBlog 𝐆𝐢𝐭𝐇𝐮𝐛 actions Azure deploy with 𝐎𝐈𝐃𝐂!
💡Over 𝟐𝟑 𝐦𝐢𝐥𝐥𝐢𝐨𝐧𝐬 secrets were exposed in #GitHub last year💀 & 𝟓𝟎K+ #Huggingface tokens leaks every month!
🛡️Switch to 𝐬𝐞𝐜𝐫𝐞𝐭𝐥𝐞𝐬𝐬 with Pipeline identity now!
👉We show you how: cloudthrill.ca/github-actio...
#Azure #NHI #CICD #Terraform #ManagedIdentity

Want to learn about @VLLm ? start here 👇🏻

🚨 As proud sponsors, we're excited to share the latest episode of #TechBeatsUnplugged! 🎧Tune in as Steve Giguere digs through every attack vector☢️on your GitHub workflows and how to protect you🛡️from them.

New Blog drop!! 👋🏻
🧠Your AI workloads are nothing without securing credentials.

🚨@CloudThrill is excited to announce its membership in the NVIDIA Inception Program! 👏🏻👏🏻👏🏻
Read full statement: cloudthrill.ca/cloudthrill-...
#NVIDIAInception Program for Startups!

TAICO May 2025 meetup!, Wed, May 7, 2025, 5:30 PM | Meetup The TAICO team is proud to announce our next meetup at the Adaptavist office in Toronto. Much thanks to [Adaptavist](https://www.adaptavist.com/ "https://www.adaptavist.com

🚨#AI & #CyberSec heads in #Toronto!
Join us on Wednesday, 𝐌𝐚𝐲 𝟕𝐭𝐡 from 5:30pm-8pm EST for another exciting #TAICO Meetup (Toronto AI and Cybersecurity Organization).
#Cloudthrill #ProudSponsor🔥
www.meetup.com/taico-toront...

Check out our team's new article how to Ace your #CNCF Certified Kubernetes Administrator exam🔥 #CKA

🧠 𝐆𝐏𝐓 𝐌𝐨𝐝𝐞𝐥𝐬 𝐟𝐨𝐫 𝐝𝐮𝐦𝐦𝐢𝐞𝐬 #cheatsheet
🤔 If you’ve opened #ChatGPT lately and thought:
“𝐖𝐚𝐢𝐭… 𝐰𝐡𝐚𝐭’𝐬 𝐨𝟑? 𝐀𝐧𝐝 𝐰𝐡𝐲 𝐚𝐫𝐞 𝐭𝐡𝐞𝐫𝐞 𝐬𝐨 𝐦𝐚𝐧𝐲 𝐦𝐨𝐝𝐞𝐥𝐬 𝐧𝐨𝐰?” You’re not alone. Today #openAI finally answered🙋🏻‍♀️
👉🏻https://platform.openai.com/docs/models/compare

Check out our team's new article how to expose you web apps demos through Zero trust 🚀 @openziti.bsky.social

📢Next week, learn more about
🧠𝐁𝐫𝐢𝐧𝐠𝐢𝐧𝐠 𝐀𝐈 𝐈𝐧𝐟𝐞𝐫𝐞𝐧𝐜𝐞 𝐂𝐥𝐨𝐬𝐞𝐫 𝐭𝐨 𝐲𝐨𝐮𝐫 𝐝𝐞𝐯𝐬: deploy 𝐀𝐈 𝐄𝐧𝐝𝐩𝐨𝐢𝐧𝐭𝐬 𝐢𝐧 𝐎𝐊𝐄 from our own🔥Oracle #ACE @clouddude.bsky.social

♠️Check out the entire agenda 👉🏻 social.ora.cl/60170pPYS
♠️ Register for free 👉🏻 social.ora.cl/60120pPYa

@oracleace.bsky.social #AIInference #K8s #ollama