Modal
Modal runs Python code on GPUs and bills by the second, with no servers to manage: you decorate functions and Modal builds the container and scales it.
- Serverless GPU
Model / inference API
Baseten
Baseten is a hosted platform for model inference and training.
Checked 2026-10-31
Loading poll...
Baseten is a hosted platform for model inference and training. Model APIs give hosted endpoints for a curated set of models, compatible with the OpenAI Chat Completions API at https://inference.baseten.co/v1. Dedicated deployments run an open model, private checkpoint or custom model on infrastructure you configure, with autoscaling, logs and metrics included. Its pricing page lists a Basic plan at $0 a month, pay as you go, with Pro and Enterprise by quote. Dedicated GPUs are billed by the minute, from $0.01052 for a T4 to $0.10833 for an H100 and $0.16633 for a B200, and Model APIs bill per million tokens. The docs say Baseten is SOC 2 Type II and HIPAA compliant. We would suggest it to teams that want to serve their own or open models without running servers. The limits are rate caps per model: an unverified Basic account gets 15 requests and 100,000 tokens a minute. The docs do not state a free tier or the free credit amount. We have not run it ourselves for this note.
Mentioned here
Runpod
Runpod rents GPUs by the second, either as pods you control or as serverless endpoints.
- GPU cloud
- OpenAI compatible
Replicate
Replicate runs models through an API and bills either by the second of hardware or by output, depending on the model.
- Model / inference API
Together AI
Together AI serves open models by the token and also rents GPUs.
- Model / inference API
- OpenAI compatible
LLM API Prices Compared: Claude, GPT, Gemini, DeepSeek
Checked 2026-10-31
How to Self-Host an LLM
Checked 2026-10-31
Modal vs Runpod: Serverless GPUs Compared
Questions
How much does Baseten cost?
Basic plan is $0 a month, pay as you go; dedicated GPUs run from $0.01052 a minute for a T4 to $0.16633 for a B200. Model APIs bill per token. Pro and Enterprise by quote (checked 31 October 2026).
Is Baseten OpenAI compatible?
Yes, it offers an OpenAI-compatible API, so most agents and SDKs can point at it by changing the base URL.
Loading poll...
Reviews
Loading reviews…
Only members can review this page. Sign in
Similar
Moonshot AI
Moonshot AI sells the Kimi API through its developer platform, now at platform.kimi.ai, and it suits teams who want Kimi K3 and K3.1 Pro with a 1,048,576-token context window straight from the maker.
- Model / inference API
- OpenAI compatible
Featherless
Featherless is a flat-rate subscription for open models, not a per-token meter, and it suits hobbyists and small teams who want to switch between many open models for one monthly price.
- Model / inference API
- OpenAI compatible
Z.AI
Z.ai sells the GLM model family through its open platform API, and it suits developers who want GLM 5.3 straight from the maker, with free small models for testing.
- Model / inference API
- OpenAI compatible
Minimax
MiniMax sells its model API through the MiniMax Open Platform, and it suits developers who want M3 or M3.1 Flash directly from the maker at a low token rate.
- Model / inference API
- OpenAI compatible
AI21
AI21 sells the Jamba model family through an API, and it suits teams that want a long-context model from a vendor that says it does not train on customer inputs.
- Model / inference API
Databricks
Databricks sells model serving inside its data platform, so it suits teams that already keep data and pipelines on Databricks and want hosted open models next to them.
- Model / inference API
- OpenAI compatible