Today, @viewyourdata announced the availability of View #AI on @OracleCloud A1 and @HPE ProLiant RL300. Featuring Ampere energy-efficiency and scalable #CPUs, the solution will support #GenAI adoption into the enterprise market. Get the details. https://t.co/DYPafF767N
Happening next week: Founder and CEO @Ampere1Renee hits the main stage at #YOTTA2024 for a conversation on sustainable #CPUs, innovations driving the industry, #AI and more with @donal888.
The Ampere team is enjoying our time at #OCW and is ready for Day 3!
We look forward to Clay Magouyrk’s #OCI keynote, kicking off at 11AM. Follow along to hear some important news...
Proud to share that @momentohq has been added as an official @awscloud event bridge API destination 🚀. It is so cool to see us added along side companies I've always looked up to like Datadog, MongoDB, and Splunk. Cant wait to see what people build!
https://t.co/PTCOxSMHBn
@RAMIGAJE@thejustinwelsh Solopreneur is defined as not having full time employees but you can have contractors or paid vendors or AI tools/services to do the menial tasks for you.
I’ve been leading a secret project for months … and the word is finally out!
🛠️ I'm proud to announce the Llama 3 Groq Tool Use 8B and 70B models 🔥
An open source Tool Use full finetune of Llama 3 that reaches the #1 position on BFCL beating all other models, including proprietary ones like Claude Sonnet 3.5, GPT-4 Turbo, GPT-4o and Gemini 1.5 Pro.
This has been months of hard work from me, my great colleagues at Groq and our awesome collaborators over at Glaive.
The model has been trained on synthetic data only. This is a powerful full finetune, not a LoRA. Yes, we've checked rigorously for overfitting using the LMSYS described robust decontamination techniques, they only score 5.6% on SFT synthetic data and 1.3% on synthetic DPO data.
Now available on the Groq API for blazing fast speeds of 1050 tok/s for 8B and 330 tok/s for the 70B model or download the open source weights from Hugging Face to start exploring & tinkering.
Check out the full blog post for more details, find all links in the reply below.
This Thursday, Ampere releases our annual update on upcoming products and milestones.
Tune in at 8AM PT to hear the latest innovations in sustainable, power efficient computing for the #cloud and #AI from @Ampere1Renee & @jwittich. https://t.co/AwL1GX4688
I was part of startups that raised billions of VC in my 20s in Silicon Valley and didn't make much money:
1. VCs tell you to dedicate your life to the business, but they take August off.
2. VCs tell you that you need to go big or go home, but their business is a collection of many bets
3. VCs tell you it's all about team, but it’s common to replace a founder/ceo.
4. VCs tell you to be lean and agile, but they’ll drown you in cash if it means a big exit.
5. VCs tell you to maintain a work-life balance, but smile when you send emails at midnight.
6. VCs tell you to own your niche, but often nudge you towards mass markets before you're ready.
7. VCs tell you to measure everything, but their decision to fund often hinges on intangible 'gut feelings'.
Am I wrong? Be honest.
Now, I don't think all VCs are the same, just like all entrepreneurs aren't the same.
But the problem isn't in VCs, it's two things:
1) Some businesses need venture funding to thrive. The SpaceX type businesses or Uber type businesses
2) VCs aren't 100% aligned with founders (especially first time founders).
Let me break it down.
They are incentivized for you to grow valuation. And sell or go public one day.
But if you get an M&A offer that might be life-changing for you and your family? They might say don't do it.
That happened to me.
I had an M&A offer that was life changing when I was 22 years old.
I would return 3x the amount I raised. Not bad right?
I brought the deal to my lead VC. And he wouldn't sign the paperwork.
"I'm not thinking big enough. Imagine what I can get in 5 years. Forget 3x, think 30x".
I'll write about that story in a future newsletter post (link in bio).
So, for most entrepreneurs, I say make your money doing bootstrapped.
Support your family. Buy your first house. Retire your parents. Do whatcha gotta do.
That "makes a ding in the universe" in a different way.
It makes a ding in your universe.
Don't be shy to share this.
Let's have an honest convo about this.
.@MomentoHQ co-founder and CTO @dymxzvf joined the Ampere devrel team for our Developer Impact Series to discuss the serverless solutions provider’s experience migrating to Ampere Altra #CPUs.
Watch their full conversation.
This is the best podcast I’ve ever done! Thanks to an awesome host @alexbdebrie, to @theburningmonk for helping me get over my fear of podcasts, and @MarcJBrooker for the inspiration for multi tenancy.
@SoftwareHuddle has the best hosts! Can’t believe they let me on!!!
Just as language models have transformed AI, CPUs are the future of AI inference. With #GPUs becoming increasingly expensive and scarce, it's hard to be free. CPU-based AI inference is only amplified now that apps and inference can run in the same place. #AI#CPUs#inference
This week, Snowflake's CFO said the company plans to save $ by using Arm-based CPUs in Azure
I wondered if he was referring to the new server chip MSFT developed itself (Cobalt), which is similar to AWS' Graviton (both Arm-based)
Turns out $SNOW is using Ampere's Arm chips!
A rare interview with @AmpereComputing CEO @Ampere1Renee ! In this 30min chat we discuss
➡ Why merchant silicon is tough
➡ Need for custom Arm cores
➡ Importance of building for customers
➡ Why silicon teams are 50%+ software
➡ AI on CPUs is necessary
https://t.co/pfnm5mt3ma