Julien from Baseten (@JulienTech) header
Julien from Baseten (@JulienTech) profile picture

Growing @Basetenco: best-in-class model hosting for AI startups. Formerly @Stripe. NYC + SF. Let's talk GPUs, sales, or Brazilian jiu-jitsu.

New York City
Joined February 2023
Photo by Julien from Baseten (@JulienTech)Photo by Julien from Baseten (@JulienTech)Photo by Julien from Baseten (@JulienTech)Photo by Julien from Baseten (@JulienTech)Photo by Julien from Baseten (@JulienTech)Photo by Julien from Baseten (@JulienTech)Photo by Julien from Baseten (@JulienTech)
Julien from Baseten retweeted
🚀 We’re excited to introduce our Speculative Decoding integration for our TensorRT-LLM Engine Builder! Our new integration allows engineers to: • Leverage SpecDec as part of our Engine Builder flow • Hit the ground running with pre-optimized configs • Lift the hood and tune for any custom use case 👉 Check out the launch blog to learn more: baseten.co/blog/speculative-… Huge shoutout to our model performance team for this feature, including @aqaderb, Justin Yi, Bryce Dubayah, and @defpan!
6
4
24
2,626
Julien from Baseten retweeted
Croissants + cloud infra really is the breakfast of champions 🥐 Had a great time hosting the AI/ML Engineering Leaders Breakfast Club this week with @baseten and @morganbarrettx in NYC.
1
1
6
810
Julien from Baseten retweeted
A video of Trump’s call with RFK Jr. was leaked - Trump talks about vaccines, tells RFK Jr. doing something with him would be “big,” mentioned his call with Biden and said the bullet was like “the world’s largest mosquito.” RFK Jr apologized for the leak.
563
3,503
1,019
9,562
4,003,612
Julien from Baseten retweeted
After witnessing so many AI builders encounter the limitations of real-time inference, we’re excited to announce async inference on Baseten! 🔄 🎉 Check out our blog to learn more: buff.ly/4cGrDr1 And don't miss our live webinar on August 15th! buff.ly/3zEyLpn
1
6
1
18
3,658
Julien from Baseten retweeted
If you’re building your own AI product, highly recommend working with @baseten Their customer support and expertise in AI infra is off the charts ❤️
3
7
2
47
8,272
Multi-model configuration is crucial for AI startups who break from the pack. Baseten's newest release makes it easier than ever. Pumped for Chains!
almost all the winning applications we’ve seen this year have been “compound ai” now delivering multi-component, performant AI systems is 10X easier! @baseten chains ⛓️ is an example of how inference growing up will enable an incredible new generation of applications
6
286
Getting coffee with an AI engineer from BrainPOP tomorrow and slowly realizing that none of my friends know what BrainPOP is. Did y'all not have a childhood!!??
1
148
Julien from Baseten retweeted
Thank you to the ML engineers, scientists, and founders who attended our NYC AI breakfast yesterday morning. New York's community of AI leaders is just getting started!
7
4
1
33
4,137
Julien from Baseten retweeted
We’re excited to launch Meta’s Llama 3 in our model library, in both 8B and 70B 🎉 The newly introduced Llama 3 is a significant improvement over Llama 2, with increased tokens, and reduced false refusal rates. These models deliver unparalleled performance, showcasing significant advancements in efficiency and speed. Our Llama 3 8B runs on A100s and Llama 3 70B runs on H100s, optimized for production.
5
21
2
221
131,485
Julien from Baseten retweeted
It is an incredibly safe assumption that models get much cheaper and smarter from here, yet very few founders are acting like it
45
30
11
498
97,085
Psyched that @mprkhrst and the @AntimetalCloud team are building such a useful and elegant product. Congrats on the launch!
Introducing the new Antimetal—a complete command and control system for AWS. The cloud is broken. It’s complex, expensive, and hard to manage. We’re solving this with AI.
3
280
Replying to @baseten
@baseten makes models go brrrrrr
Baseten launches Mistral 7B API with leading performance 🚀 @baseten has entered the arena with their first serverless LLM offering of Mistral 7B Instruct. Artificial Analysis are measuring 170 tokens per second, the fastest of any provider, and 0.1s latency (TTFT) in-line with the next fastest provider. Mistral 7B Instruct from @Mistralai has become the leading 7B-class model for use-cases from RAG chatbots to data extraction. With pricing more than two orders of magnitude lower than large models like GPT-4 and Claude 3 Opus, models like Mistral 7B Instruct enable use-cases that wouldn’t be possible with the economics of larger models.
5
90
Replying to @baseten
@baseten's just getting started. Couldn't be prouder to work for this brilliant team with insane potential.
6
88
Julien from Baseten retweeted
I have used a LOT of GPU providers. Hands down the most reliable one is @baseten They never have gone out for us, the team is always around, and the Truss system just…. makes sense…
3
18
2,499
Need H100s? DM me
Launching today 🎉 Double your throughput or halve your latency for @MistralAI, @StabilityAI + others? Do both at ~20% lower cost with @nvidia H100s on Baseten. Here’s how 👇
1
5
293
Replying to @theesabina
obsessed w this artist
1
1
86
Replying to @mschoening
@mschoening this is sick. We'd love to know what you think of @baseten and get your feedback. Let me know if you want us to send over free credits. Excited to hear how the test goes!
54