News icon

Kimi K3 is now available on Runpod

G24.6

The AI Developer Cloud

Experiment, train, fine-tune, deploy, scale on one platform

1M+ developers on Runpod, and the cloud we're building next.

Updated September 27, 2026
1 million developers on Runpod

What’s new

Performing your first LLM fine-tune on Runpod with Axolotl, and how to host it after

October 12, 2026

You can teach an open-weight LLM to text like a specific persona: their slang, their reply length, their humor, even how they write in a roleplay scene, creative writing exercise, or chatbot customer interactions. All it takes is some chat logs (synthetic or real), a few hours of data prep, and a single GPU pod. This guide walks through the whole process on Runpod with Axolotl, from raw messages to a model you can chat with in SillyTavern, KoboldCpp, or text-generation-webui, using two base models: a 27B Qwen model and the 128B Mistral Medium 3.5.‍ Everything here was tested end to end, and every error we hit along the way is in the Troubleshooting section, so you shouldn't get stuck where we did.

Read blog

Global Volumes are now in beta for Serverless endpoints

October 8, 2026

Attach a Global Volume to a Runpod Serverless endpoint and your workers start in any data center with your GPU types. Store model weights once, read anywhere.

Read blog
Cover of the Fall 2026 State of AI Compute report

Fall 2026 State of AI Compute

Insights from the latest data around AI deployment, infrastructure demand, and model scaling trends.

Solution

One platform. Full lifecycle.

Go from experiment to production without replatforming. Pods, Serverless, and Clusters, all in one account.

Launch a GPU pod in seconds.

Spin up a fully-loaded, GPU-enabled environment in under a minute. From B200s to RTX 4090s, Runpod supports over 30 GPU SKUs.

Deploy globally with a few clicks.

Run workloads across 31 global regions worldwide, with low-latency performance and global reliability.

Scale on autopilot with Serverless.

From 0 to thousands of compute workers, adapting to your workload in real time, only paying for what you use.

Runpod console GPU pod launch visual
Zero to production workflow diagram

Go from experiment to production in one flow.

One account. No migrations between stages.

  • GPU environment launch workflow illustration

    Spin up

    GPU environment in under 30 seconds. 30+ GPU SKUs, 31 global regions.

  • AI model training workflow illustration

    Build

    Train models, fine-tune, process data. Your containers, your framework, your code.

  • Serverless endpoint deployment workflow illustration

    Deploy

    Write your handler. Push to Serverless. Live inference endpoint, auto-scaling, zero idle cost.

  • Scaling workers workflow illustration

    Scale

    0 to hundreds of concurrent workers in under 250ms.

Enterprise grade uptime.

Runpod handles failovers, ensuring your workloads run smoothly, even when resources don’t.

Managed orchestration.

Runpod Serverless queues and distributes tasks seamlessly, saving you from building orchestration systems.

Real-time logs.

Get real-time logs, monitoring, and metrics with no custom frameworks required.

Features

Production inference without the warm-up tax.

Most serverless GPU options make you choose: pay for idle capacity, or eat cold-start latency. Runpod Serverless does neither. Scatter Lab handles 1,000+ inference requests per second on Runpod, at nearly half the cost of major cloud providers.

Autoscaling configuration visualPurple glow background

Autoscale in seconds

0 to thousands of workers. Automatically. No config files.

FlashBoot serverless settings visualPurple glow background

Sub-200ms cold starts

FlashBoot eliminates warm-up engineering. Sub-200ms.

Always-on worker setting visualPurple glow background

Zero idle cost

Your endpoint costs nothing when it's not running.

Persistent network storage illustrationPurple glow background

Persistent network storage

Full AI pipelines, no egress fees.

Case Studies

In production. At scale.

See what our customers are building.
How Rendair scaled to 13M AI-generated images on Runpod

"We didn't seriously evaluate other alternatives before committing to Runpod. We found their approach interesting from the beginning, particularly the ease of integration, and we haven't had a reason to switch since."

How Nodalview chose what to move

"AWS is very reliable and has everything we need, but it is way more expensive than Runpod. With Runpod, we can choose the right card for the job."

How Faceless.video bootstrapped to $1M+ ARR

"Runpod is easily one of the most scalable tools that we use in our infrastructure."

How BottoDAO Cut Its AI Art Engine's Costs by 73%

"What's happened is, moving to Runpod, Botto has a better GPU, better graphic models and better resources, but it also costs less. That's something I didn't expect we were going to have."

How Open Source Films built an open-source VFX and agentic development pipeline on Runpod

"You guys have genuinely changed the scope of what we've been able to attempt."

How TOOL Scales Big AI Ideas on Runpod

"All of these projects, the renders for AMD, the Coca-Cola builds, that has to do with scalability. If we can't scale, we can't deliver. Runpod makes that possible."

How Aneta Handles Bursty GPU Workloads Without Overcommitting

"Runpod has changed the way we ship because we no longer have to wonder if we have access to GPUs. We've saved probably 90% on our infrastructure bill, mainly because we can use bursty compute whenever we need it."

How Gendo uses Runpod Serverless for Architectural Visualization

"Runpod has allowed the team to focus more on the features that are core to our product and that are within our skill set, rather than spending time focusing on infrastructure, which can sometimes be a bit of a distraction.”

How Civitai Trains 800K Monthly LoRAs in Production on Runpod

"Runpod helped us scale the part of our platform that drives creation. That’s what fuels the rest, image generation, sharing, remixing. It starts with training."

How Scatter Lab Powers 1,000+ Inference Requests per Second with Runpod

"Runpod allowed us to reliably handle scaling from zero to over 1,000 requests per second in our live application."

How InstaHeadshots Scales AI-Generated Portraits with Runpod

"Runpod has allowed us to focus entirely on growth and product development without us having to worry about the GPU infrastructure at all."

How KRNL AI scaled to 10K+ concurrent users while cutting infra costs 65%.

"We could stop worrying about infrastructure and go back to building. That’s the real win.”

How Coframe scaled to 100s of GPUs instantly to handle a viral Product Hunt launch.

“The main value proposition for us was the flexibility Runpod offered. We were able to scale up effortlessly to meet the demand at launch.”

How Glam Labs Powers Viral AI Video Effects with Runpod

"After migration, we were able to cut down our server costs from thousands of dollars per day to only hundreds."

How Segmind Scaled GenAI Workloads 10x Without Scaling Costs

Runpod’s scalable GPU infrastructure gave us the flexibility we needed to match customer traffic and model complexity, without overpaying for idle resources.

Impact

Evaluate GPU infrastructure by workload fit.

Compare GPU availability, deployment workflow, pricing model, support path, and capacity planning before choosing a platform.

Enterprise

Enterprise-grade from day one

Built for scale, secured for trust, and designed  to meet your most demanding needs.

Reliability and uptime illustration

99.9% Uptime

Run critical workloads with confidence, backed by industry-leading reliability.

Security and compliance illustration

Secure by default

Independently audited SOC 2 Type II compliance for end-to-end data protection.

Scalable global infrastructure illustration

Scale to thousands
of GPUs

Adapt instantly to demand with infrastructure that grows with you.

FAQs

Questions? Answers.

Serverless, simplified. Clear answers on running your code without the fuss.

Runpod offers three primary infrastructure products: Serverless (autoscaling GPU endpoints that scale to zero when idle), Pods (GPU instances for persistent compute and development – available as Reserved (guaranteed) or Spot (interruptible, lower price), and Clusters (multi-GPU distributed compute for training and large-batch inference). All run on the same GPU catalog (H100 80GB HBM3, H100 NVL, A100, L40S, and more) accessible on-demand with no contracts or minimum commitments. See pricing

AI IaaS is on-demand, cloud-based access to GPUs, networking, and storage; you rent the capacity you need by the hour or second instead of buying and operating the hardware yourself. The tradeoff is the same one cloud computing has always offered: building your own data center gives you full control and, at very high sustained utilization, a lower per-hour cost, but it requires upfront capital, lead time to provision, and an ops team to keep it running. AI IaaS platforms like Runpod let you start serving traffic in minutes, scale to zero when there's no demand, and avoid being locked into hardware that ages out as new GPU generations ship.

AI agent infrastructure is the compute, storage, and networking layer that AI agents depend on to run, call tools, store memory, and scale. Runpod supports it across three layers: Serverless endpoints for fast inference tool calls, persistent Pods for stateful agents that need to stay running, and network volumes for shared memory and model weights across workers. The Runpod skills package lets Claude Code, Cursor, and other coding agents deploy and manage Runpod resources directly.

For inference at scale, Runpod Serverless gives you autoscaling GPU endpoints with sub-200ms cold starts (powered by FlashBoot) and a built-in job queue across 31 global regions. For training and fine-tuning at scale, Clusters support 200+ simultaneous GPUs with InfiniBand. For teams with compliance requirements, Secure Cloud provides network-isolated environments. Most teams use all three depending on the stage of their workload: Serverless for production inference, Pods for development and agent hosting, Clusters for training runs.

Yes. Runpod is SOC 2 Type II certified and HIPAA and GDPR compliant, with SOC 2 reports, BAAs, and DPAs available for security review through the Runpod Trust Center. Secure Cloud adds network isolation for workloads with stricter compliance needs. Coverage can vary by region and deployment model, so check requirements for your workload. Enterprise agreements add a 99.99% uptime SLA, dedicated capacity, and tailored terms. See runpod.io/legal/compliance for current certification details.

The AI developer cloud.

No replatforming. No lock-in. No hyperscaler tax. Ready when you are.

Star field background