Podcode
Sign in
Providers

Every model.
Every host. One UI.

Run Qwen3-Coder on a GPU account you control for a price you agree up front. Deploy your private model in minutes, pause it when you're done, and resume it in a click - it only bills while it's serving. Run a quantized Llama on your laptop when a single token can't leave your machine. Cloud APIs stay available if you want them, and when a lab reprices you can swap without re-tooling.

A
Anthropic
Frontier cloud models via API.
API key
O
OpenAI
Frontier cloud models via API, or use your ChatGPT subscription.
API key · Subscription
G
Google AI Studio
Google’s cloud models via API, or use your Code Assist plan.
API key · Subscription
V
Google Vertex AI
Managed cloud models, plus customer-tuned models.
Service account
B
AWS Bedrock
Managed models inside your own AWS account and region.
IAM role · access key
M
Mistral La Plateforme
Mistral’s cloud models via API.
API key
C
Cohere
Cohere’s cloud models via API.
API key
OR
OpenRouter
One key for Claude, GPT, Gemini, GLM, and the rest of the long tail.
API key · OAuth
F
Fireworks AI
Open-weight models on serverless endpoints, billed per token.
API key
N
NVIDIA NIM
Open-weight models behind one OpenAI-compatible endpoint.
API key
Z
Z.AI
GLM family (glm-4.6, glm-4.5) over an OpenAI-compatible endpoint. Coding-plan subscription keys supported.
API key · Subscription
Mx
MiniMax
MiniMax-M3 and M2.x family over an Anthropic-compatible endpoint. Coding-plan subscription keys supported.
API key · Subscription
GH
GitHub Copilot
Use your GitHub Copilot subscription for GPT, Claude, and Gemini models.
Subscription

Run on what you trust.

Cloud when you want frontier intelligence, local when a token can't leave the building, and your private pod only costs you while it runs - and Podcode never sits in the billing or inference path. Your providers, your spend, your walls.