Friendli Model APIs
Serverless, OpenAI-compatible API for pre-optimised open-source frontier models on the Friendli inference stack, with tool calling and JSON mode, billed per token.
Find alternatives to Friendli Model APIs
What it does
- Model inference
Sources
- Pricing
-
Pricing page, Model APIs table: "Pay per token" with a "$ / 1M tokens" column per model.
- Description
-
Official documentation · 16 Nov 2026
Page quote: "Run the fastest frontier model inference with a simple API call. No infra to manage."
- What it does
-
Official documentation · 16 Nov 2026
Page quote: "Run the fastest frontier model inference with a simple API call. No infra to manage."
- Sold within
-
Pricing page: "Get started instantly with self-serve, or contact us for enterprise deployments." and a Model APIs per-token price table.
- Name
-
Official documentation · 16 Nov 2026
Page quote: "Run the fastest frontier model inference with a simple API call. No infra to manage."
- URL
-
Official documentation · 16 Nov 2026
Page quote: "Run the fastest frontier model inference with a simple API call. No infra to manage."
- Deployment
-
Official documentation · 16 Nov 2026
Page quote: "Run the fastest frontier model inference with a simple API call. No infra to manage." The pages read describe a vendor-run web service and no other deployment model, so saas is the vendor-operated reading.
- Interfaces
-
Official documentation · 16 Nov 2026
Product page: "Run the fastest frontier model inference with a simple API call."
Also from FriendliAI
-
Friendli Dedicated Endpoints API service
Friendli Dedicated Endpoints deploy open-weight or custom models on dedicated GPU instances, billed by GPU time or reserved by the month.
-
Friendli BYOG Infrastructure service
Runs the Friendli inference stack on the customer's own cloud GPUs (AWS and GCP), with FriendliAI deploying, operating and scaling it.
Open FriendliAI in the directory
Something wrong here? Send a correction — quote this product id: friendliai-friendli-model-apis.