Runpod
@runpod
Followers
7K
Following
833
Media
131
Statuses
604
GPU Cloud built for production https://t.co/PWVZocfp6D
Joined March 2022
We've shipped RunPod Instant Clusters - scale to 64 H100 GPUs across 8 nodes with zero setup. The ability to train massive models (720B+ parameters) should be accessible to everyone building the future. No sales calls. No contracts. Just build.
8
10
76
Cached models are now GA for Runpod Serverless endpoints. Cached models unlock: ⚡Faster cold starts (when the model is already on the host) ⚒️More predictable deployments (no “start worker then wait on downloads”) 🤏Smaller images (app + runtime, not weights) 📦Less redundant
2
0
10
Try Chatterbox Turbo for free this week → https://t.co/Zp1vkpH2hx
console.runpod.io
Resemble AI's Chatterbox Turbo is the fastest open source text to speech model with paralinguistic tags for non-speech sounds with expressive voice cloning capabilities.
Introducing Chatterbox Turbo, the fastest open source Voice AI model with emotions. Our gift to the dev community this holiday season! • ~6x faster RFT • Expressive sound tags: sighs, laughs, coughs • PerTh watermarking on every output Here's everything you need to know 👇
0
0
3
The best open-weight LLM by a US company, Cogito v2.1 671B, was trained on Runpod as part of their frontier stack. 🇺🇸🦅🚀 Try it now with no setup:
console.runpod.io
The Cogito v2.1 LLMs are instruction tuned generative models. It has been optimized for coding, STEM, instruction following and general helpfulness. Cost is $0.5 per 1M tokens.
Today, we are releasing the best open-weight LLM by a US company: Cogito v2.1 671B. On most industry benchmarks and our internal evals, the model performs competitively with frontier closed and open models, while being ahead of any US open model (such as the best versions of
0
3
11
As an AI founder, the last thing on your mind should be worrying about compute. Let us handle that. Autoscaling? We do that. Sub 500ms coldstarts? That's us. RTX 5090s for speedy inference? Got that too. Pay only for what you use --- try it👇 https://t.co/C3gLlxb6B3
runpod.io
Skip the infra headaches. Our auto-scaling, pay-as-you-go, no-ops approach lets you focus on building. Pay only for the resources you consume, billed by the millisecond.
1
0
5
it was unreal to meet @DesireeCachette, @danielhanchen + michael from @UnslothAI and @calcsam + @abhiaiyer from @mastra_ai irl huge shout out to @max4c_ for putting together an awesome guide on how to get started with your own agent: 1. fine-tune with @UnslothAI 2. upload to
1
5
12
Our API services have mostly recovered. We are currently experiencing a temporary GPU shortage due to a backlog of requests during the outage. The system will need some time to process the queued jobs. The team is actively monitoring the situation.
3
1
7
We're beginning to see early signs of recovery in the affected AWS region, and parts of our system are coming back online. The team is closely monitoring the situation and will continue working on the migration as a precaution. AWS health Status:
0
0
3
Status update: Runpod is impacted by the AWS us-east-1 outage. The Runpod console currently won't load however • Your Pods are still running. • Pods will not be terminated. • You are not being billed for affected services. • Serverless endpoints cannot receive new requests.
3
2
16