Modal
@modal
Followers
19K
Following
3K
Media
212
Statuses
1K
Bring your own code, and run CPU, GPU, and data-intensive compute at scale. The serverless platform for AI and data teams.
New York City
Joined July 2022
Honored to be highlighted as a top AI enabler in @MadronaVentures's IA40 list!. Lovely to see several of our customers on this list, too ;).
The 2025 Intelligent Applications 40 is here. Now in its fifth year, the #IA40 recognizes the top private companies shaping the future of applied AI, from foundational infrastructure to full-stack intelligent applications. This year’s list reflects the scale and speed of
0
0
9
Modal customers always do it big. Millenia of audio next?. Read the full case study here:.
modal.com
Zencastr scaled up to 1,500 concurrent GPUs on Modal to process hundreds of years of podcast audio in just a few days. Today they run transcription, speaker detection, and audio enrichment for...
0
0
1
Zencastr's existing Kubernetes set-up wasn't a good fit for GPU-based models. 1. It couldn't scale up and down in a cost-efficient way. 2. It was time-consuming to keep different models, frameworks, and GPU drivers in sync. Modal made it easy to set up spiky, multi-model
1
0
1
RT @charliermarsh: We’re cohosting an event with @modal here in NYC on September 10!. Come meet folks from the team and learn more about ou….
0
5
0
Try it out! Modal makes it easy to set up the right environment for Unsloth and then launch your job without thinking about infrastructure.
modal.com
Training large language models is an incredibly compute-hungry process. Open-source LLMs often require many GBs (or in extreme cases, one TB!) of VRAM just to fit in memory. Finetuning models...
1
1
9
Modal x @UnslothAI: new tutorial on efficient LLM fine-tuning 🦥. Unsloth is a popular library for performant fine-tuning on single GPUs. You can get 2x faster training with 70% less memory usage compared to Flash Attention 2.
1
2
15
RT @bernhardsson: Two of the smartest people I know going super deep on AI infra and @modal's secret sauce!. @akshat_b (CTO and cofounder o….
0
9
0
RT @zavaindar: our most cutting edge companies consistently use @modal . this was such a great and inspiring watch of real disruptive infra….
0
2
0
we 💚 supporting open-source projects!.
The @DeepSpeedAI would like to thank @modal for sponsoring our gpus for CI. This is an amazing contribution to our AI-democratizing open source project. The Modal team is outstanding in their amazing support - speed, expertise and a human experience!.
0
0
9
When existing primitives aren't fast enough for your vision, you have to build them from scratch. Co-founder/CTO @akshat_b goes deep with @AmplifyPartners on building AI infra from the ground up 🏗️.
At @modal we've built every layer of the AI infra stack from scratch — from filesystems and networking to our own async queues and multi-cloud GPU orchestration. I sat down with @narayanarjun from @amplifypartners to go into depth on all of this, including the fun ways the
0
2
21
👀 The most in-depth look at our technology to date. If you're curious how Modal is so damn fast, check out this blog post from @AmplifyPartners!.
What does @modal do? How does it work? What's different about AI infra? Why did we throw out Kubernetes and Docker built our own infra stack from scratch? . @AmplifyPartners wrote this article about a lot of the gory details under the hood of Modal – link in 🧵.
1
4
38
RT @krandiash: we have a lot of modal enjoyers @cartesia_ai, really incredible company and people.
0
2
0
It’s been exciting to help support @cartesia_ai on Line, their new code-first voice agent platform — solving how they instantly cold start voice agents at scale. ⚡️. Here’s a fun one we built that tells you how to get started on Modal - call +1(626)746-1433 🤙
Introducing Line by Cartesia: the modern voice agent development platform. Line was built to be code-first, because best-in-class products are built in code. ▶️ Watch us build an advanced voice agent with background reasoning in just minutes.
0
6
78
Get the full download here:.
modal.com
Welcome to another round of Modal Product Updates! Here's what's new this month.
0
0
4
🚀 New in Modal this month: GPU memory snapshots (10x faster cold boots), notebooks, service tokens, uv support, gpt-oss + vLLM tutorial, and more!.
7
2
108
Modal makes it easy to test and launch training runs without configuring and managing cloud infra. Give it a try!
modal.com
This example demonstrates how to fine-tune an ASR model (whisper-tiny.en) and deploy it for inference using Modal.
0
0
2
We tried this out on a "Science and Technology" dataset with 7k samples:. ✅ Relative Word Error Rate improvement of 12%.⏱️ 1.5 hour training run.💰 Cost only $12 on Modal.
1
0
2
New tutorial on Fine-tuning Whisper 🎙️. Open-source ASR models are having a ~moment~. You can fine-tune them to achieve higher accuracy on things like domain-specific vocab.
1
2
23
RT @therealprady: we generate thousands of hours of video everyday with 99.9% uptime on infra we built in <6 months. these aren’t OSS / eas….
0
3
0