TensorPool

Hi all! We’ve shipped container support and an updated version of our Jobs product! Container Support: Users can now select from base images to be deployed on their clusters Jobs: You can now submit a job directly to an existing cluster Let us know what you think :D

Pretty proud of this cracked Jupyter NB for ML we've been working on! It solves the reproducibility issues, flaky kernels, and inability to scale experiments that you see with traditional notebook paradigms. Check out this demo! youtu.be/LIQqzb0QGw8?si…

@AcceleratedCS @rrhoover we gotchu!

@rrhoover Cc @TensorPool

Request for GPUs: A portfolio co is scaling fast and needs 400 H100s or 275 B200s asap for a one week reservation (potential to convert to a longer contract). LMK if you have a hookup. DMs open.

vibe training is here!!! watch me easily fine-tune glm-4 using CC & TensorPool. SKILL.md used: skills.sh/tensorpool/ten… Model fine-tuned: huggingface.co/zai-org/glm-4-… youtu.be/BUDCZYcpv0Q?si…

@TensorPool Takeover

After launching SongRater — our open platform for music perception data 💙 — we’re excited to share a big update: Tensorpool is sponsoring $20 for everyone who annotates at least 2 songs! Open music AI needs open data. You can help → songrater.bud-e.ai

RT @TychoSvoboda: It’s always amazing to see small teams outperform companies with $100M+ in funding, and even more amazing when you get to…

B300s have entered the pool at tensorpool.dev! Starting at $5.49/hr.

At TensorPool, we've worked with world-class research teams and startups training foundation models. We've noticed they all face the same three challenges: 1) Underutilization and idle time is inevitable and expensive 2) Researchers see model training as a set of experiments 3) Model developers are stuck on statically sized clusters We built TensorPool Jobs to fix this. youtu.be/EAqMb4GcV0c?si…

More and more people are choosing B200s over H100s. We did the math on why. Read here: tensorpool.dev/blog/gpu-price…

For multi-node training jobs, storage speeds often become the bottleneck. Read how TensorPool solves that. tensorpool.dev/blog/fast-nfs-…

AMD Instinct MI300x have entered the pool on tensorpool.dev! Starting at $2.49/hr. #AMD #MI300x

Excited to launch our tutorial series. Learn how to easily spin up a GPU and attach storage. youtu.be/4qPY2UHG71o

Multi-node B200s and H200s have entered the pool on tensorpool.dev! B200s starting at $4.99/hr, H200s starting at $2.99/hr.

We're hosting the largest applied AI research workshop + hackathon. May 23-24 @AGIHouseSF -> agenthacks.org $10K in prizes, $25K in credits, tech talks, bounties. Hosts @dexterity_ai, @afterquery, @agihousesf. If you're a top student, researcher, or builder - DM.

Just dropped: - P4s - P100s - V100s $5 free credits at tensorpool.dev

New TensorPool feature alert! You can now train a job on multiple GPUs.

@Gurpret_ It’s $1.51/GPU/hr!

@TensorPool What’s the pricing like?

To celebrate our launch we just dropped A100 80s! Get started today at tensorpool.dev

🌊@TensorPool is the easiest way to use cloud GPUs. They bring the world’s GPUs to your local IDE. ycombinator.com/launches/Mq0-t… Congrats on the launch, @TychoSvoboda, @joshkmartinez, and @hlumelonotshe!