“We use Modal as our sandbox provider for executing code within Lightfield chats.”
Open Model Hosting & GPUsMaker says so · source ↗
Modal
Cloud platform for running Python functions, inference, training and sandboxes on serverless GPU/CPU infrastructure.
60 makers' products we track use Modal. For open model hosting & GPUs it's 2nd most chosen of 9 options: 60 of the 185 makers' products we know the open model hosting & GPUs of (32%). Among the 48 launched in 2026, 48% chose it, against 30% of those launched in 2024 or earlier. 22 open-source projects also declare it in their code.
Profile checked
How makers use Modal
Makers' products we track use Modal for open model hosting & GPUs (60), most often alongside Vercel (38%), Next.js (32%), Resend (22%).
What Modal is used for
Modal vs the alternatives makers chose
For open model hosting & GPUs, 60 of the 185 makers' products we know the choice of picked Modal (32%), behind Hugging Face (42%). Among those launched in 2026 it's 48%.
| Option | Makers' products | 2026 launches | Open source | |
|---|---|---|---|---|
| 77 42% | 52% | No | ||
| 60 32% | 48% | No | ||
| 23 12% | 2% | No | ||
| 19 10% | 2% | No | ||
| 12 6% | 0% | No | ||
| 5 3% | 2% | No |
For open model hosting & GPUs: how many of the 185 makers' products we know the open model hosting & GPUs of chose each option, and the share among the 48 launched in 2026, since a tool that came first has products from every year.
Often paired with
Share of the 60 makers' products on Modal that show this choice.Markets that build with it most
Product Hunt markets where Modal is in far more of the stacks we know than across every market (1%).
When to use Modal, and when not
Pick Modal if
- Your workload is your own Python code — a pipeline, batch job or custom model — that should scale to zero. Open Model Hosting & GPUs →
- You already deploy on Vercel or Modal and want sandboxes on the same bill. Code Sandboxes for Agents →
Something else fits better if
- The open model you need is already served by a host and you want to pay per token: Together AI or Fireworks AI or Hugging Face.
- You want a raw GPU machine for experiments, training or a long-running server: Runpod.
- One model is core to the product and needs steady latency under real traffic: Baseten or Hugging Face.
- You run the GPUs yourself and need an OpenAI-compatible server for many concurrent users: vLLM.
- Your agent analyses data or runs code it wrote, one disposable computer per session: E2B or Daytona.
How Modal itself is built
6 tools, from its own code, website and Product Hunt page.
Who uses it
60 makers' products, each linked to the source that shows it, and 22 open-source projects that declare it in their code. The best known on Product Hunt: viktor.com (2,441 followers), Lightfield (2,260 followers), Cekura (2,220 followers), Gan.AI (1,844 followers), Kodezi (1,733 followers). 2 of the makers' products are gone now (domain lapsed, up for sale or taken down); they chose Modal while live.
“We run large training and inference jobs on Modal!”
Open Model Hosting & GPUsMaker says so · source ↗“Modal helps us spin up Python workloads with speed and control. Great for agents that need to run fast, ephemeral jobs. A dream for short-lived compute without DevOps overhead.”
Open Model Hosting & GPUsMaker says so · source ↗“Modal lets us run isolated, on-demand ingestion workers so large source imports stay fast and dependable.”
Open Model Hosting & GPUsMaker says so · source ↗“We use Modal to help with scalable GPU serving for our models.”
Open Model Hosting & GPUsMaker says so · source ↗“Our Modal training pipelines support our router customization workflows.”
Open Model Hosting & GPUsMaker says so · source ↗“Modal offers really great reliability and scale, as well as several first-class features that other sandbox providers don't have.”
Open Model Hosting & GPUsMaker says so · source ↗“Best in class GPU containerization experience. So so nice having GPU autoscaling and containerized deploys out of the box.”
Open Model Hosting & GPUsMaker says so · source ↗“the easiest way to serve on B200”
Open Model Hosting & GPUsMaker says so · source ↗“Helps us run intensive AI models without worrying about cold starts or scaling.”
Open Model Hosting & GPUsMaker says so · source ↗“Solving inferencing for us at scale.”
Open Model Hosting & GPUsMaker says so · source ↗“Super easy to spin up and scale my API requests. Would highly recommend anyone using this.”
Open Model Hosting & GPUsMaker says so · source ↗“Serverless GPU infrastructure helps us scale to workloads of hundreds of H200s/B200s within seconds to maintain low latency for our retrieval models in a cost-effective way.”
Open Model Hosting & GPUsMaker says so · source ↗“Viktor runs its agents server-side, 24/7, which means spinning up isolated sandboxes for thousands of teams at once and scaling back to zero the moment they go idle. We looked at rolling our own on Kubernetes and at the usual serverless suspects. Modal was the only one where a Python engineer could define the whole sandbox, schedule, and scaling story in code and ship it the same afternoon. It is a big part of why a small team can run this much compute without standing up a platform team to babysit it.”
Open Model Hosting & GPUsMaker says so · source ↗“Can't imagine deploying Python apps any other way”
Open Model Hosting & GPUsMaker says so · source ↗“Modals VM's power our AI agent environments, allowing them to view pull requests, inspect files, and even execute code! Thanks to the Modal team for prioritizing our feature requests and building a GREAT headless compute environment.”
Open Model Hosting & GPUsMaker says so · source ↗“Modal has been a backbone tool for the AI team at Fathom.”
Open Model Hosting & GPUsMaker says so · source ↗“Modal's on-demand containers are a natural fit. Same composable, no-infra ethos as mngr itself.”
Open Model Hosting & GPUsMaker says so · source ↗“Modal helps us to deploy services effortlessly, and host backend infrastructure without any overheads. A beautiful DX.”
Open Model Hosting & GPUsMaker says so · source ↗“Modal gives us fast spin up times, simple serverless workflows, and a clean developer experience. It let us prototype Kodezi pipelines without managing GPUs or infra overhead. The combination of speed, scale, and low friction made it easier to iterate than the alternatives.”
Open Model Hosting & GPUsMaker says so · source ↗“We use Modal for both model training and inference. 10/10 recommend! Amazing product and customer support”
Open Model Hosting & GPUsMaker says so · source ↗“Power our agentic search tool for restaurants and bars discovery.”
Open Model Hosting & GPUsMaker says so · source ↗“Amazing simplicity and cost efficiency”
Open Model Hosting & GPUsMaker says so · source ↗“runs our GPU compute. fast cold starts and clean serverless primitives made the deploy layer possible”
Open Model Hosting & GPUsMaker says so · source ↗“Handles the heavy lifting for our training and generation pipelines without friction. Modal’s simplicity and speed make it perfect for scaling AI workloads like real-time photo generation.”
Open Model Hosting & GPUsMaker says so · source ↗“Easy to set up, great for payment only when using services. Custom transcription and diarization pipelines.”
Open Model Hosting & GPUsMaker says so · source ↗“The easiest way to host Python especially when you need to do some AI stuff.”
Open Model Hosting & GPUsMaker says so · source ↗“Secure sandboxes to execute arbitrary code”
Open Model Hosting & GPUsMaker says so · source ↗“Modal helped us train our model to generate human-like content, we developed state-of-the-art content engine that writes content indistinguishable from the one written by subject-expert human.”
Open Model Hosting & GPUsMaker says so · source ↗“Modal provides the best developer experience out there for AI startups shipping quickly!”
Open Model Hosting & GPUsMaker says so · source ↗“Compute infrastructure that's easy to work with”
Open Model Hosting & GPUsMaker says so · source ↗“Modal was a complete game-changer for FlexChords. I run heavy PyTorch audio models and Demucs stem separation on serverless GPUs with zero cold-start headaches. Deploying containerized ML pipelines from Python in seconds without managing Kubernetes clusters made my backend possible.”
Open Model Hosting & GPUsMaker says so · source ↗“Highly scalable model serving infrastructure”
Open Model Hosting & GPUsMaker says so · source ↗“I chose Modal because it made running data-heavy workloads feel effortless. A lot of alternatives required more setup or infrastructure thinking, but with Modal I could just focus on the logic and let it handle the scaling and execution. It’s been especially useful for processing environmental data in the background without worrying about servers. It strikes a really nice balance between power and simplicity, which made building and iterating on My!hū much smoother.”
Open Model Hosting & GPUsMaker says so · source ↗“We use Modal to serve TTS and embedding models with scalable GPU power”
Open Model Hosting & GPUsMaker says so · source ↗“Simplified building and scaling the Accent Oracle web app with serverless infrastructure.”
Open Model Hosting & GPUsMaker says so · source ↗“Huge shoutout to Modal! The cloud infrastructure made spinning up AI and ML tasks for ScaryStories.Live effortlessly. We couldn’t have scaled our real-time horror experiences without you. We’re truly grateful for your generous support in our early days!”
Open Model Hosting & GPUsMaker says so · source ↗“We use a number of AI models that are run on Modal infrastructure.”
Open Model Hosting & GPUsMaker says so · source ↗“Monthly credits helped me make this product a reality”
Open Model Hosting & GPUsMaker says so · source ↗“Shoutout to Modal for running our models. Outpainting full films takes serious GPU horsepower, and Modal lets us scale that compute up and down without ever thinking about servers. It's been a game changer for us.”
Open Model Hosting & GPUsMaker says so · source ↗“Ease of use and familiarity!”
Open Model Hosting & GPUsMaker says so · source ↗“We run much of our backend on Modal, which easily lets us fan out requests.”
Open Model Hosting & GPUsMaker says so · source ↗“All of SEOWidget's widgets are powered by Modal. It's extremely easy to set up FastAPI endpoints on them.”
Open Model Hosting & GPUsMaker says so · source ↗“For when you want to quickly spin-off a fine-tuned model from AnythingLLM. There is no easier way to build that pipeline we offer in-app than using Modal Labs for creating fine-tuning jobs on powerful GPUs for a fair price.”
Open Model Hosting & GPUsMaker says so · source ↗“We run our models on Modal. Amazing DX, brilliantly documented API, and very fast and helpful support on their Slack.”
Open Model Hosting & GPUsMaker says so · source ↗“We use Modal to run all of Pioneer's fine-tuning and inference workloads. We evaluated running our own GPU infra on AWS and tried a few other serverless GPU providers, but Modal was the only one where spinning up an H100 job felt as easy as calling a Python function.”
Open Model Hosting & GPUsMaker says so · source ↗“Helps us serve Reducto at scale with large workloads”
Open Model Hosting & GPUsMaker says so · source ↗None of these in makers' own words.
Open source: a project that declares Modal as a dependency in its public code — verifiable, but not necessarily a live product.
Alternatives to Modal
All alternatives by situation →Questions makers ask about Modal
Is Modal good for production?
60 makers' products we track run on it, and 2 of them have since gone (the product, not necessarily because of Modal).
What is Modal used for?
Cloud platform for running Python functions, inference, training and sandboxes on serverless GPU/CPU infrastructure. Makers' products we track use it for open model hosting & GPUs (60).
What are the alternatives to Modal?
For open model hosting & GPUs, the other options makers' products we track chose most are Hugging Face (77), Baseten (23), Runpod (19), Together AI (12), Fireworks AI (5). Open source or self-hostable among them: Baseten. Each alternative, with the situation it fits, is on the Modal alternatives page.
Is Modal still popular with new products?
More than before: among the 48 makers' products launched in 2026 that made the open model hosting & GPUs choice, 48% chose Modal, against 30% of the 46 launched in 2024 or earlier.
When should you use Modal?
Your workload is your own Python code — a pipeline, batch job or custom model — that should scale to zero (open model hosting & GPUs). You already deploy on Vercel or Modal and want sandboxes on the same bill (code sandboxes for agents). Something else fits better if: the open model you need is already served by a host and you want to pay per token, pick Together AI or Fireworks AI or Hugging Face; you want a raw GPU machine for experiments, training or a long-running server, pick Runpod.
Is Modal open source or self-hostable?
Not open source, and hosted only.
Can AI coding agents work with Modal?
No llms.txt, official MCP server or CLI found yet.
Who uses Modal?
60 makers' products we track, each with a source, and 22 open-source projects declare it in their code. The best known of them on Product Hunt: viktor.com (2,441 followers), Lightfield (2,260 followers), Cekura (2,220 followers), Gan.AI (1,844 followers), Kodezi (1,733 followers). source ↗