Isoquant
Inference API for open models with a chat-completions interface: GLM-5.3-Flash at $0.09 per million input tokens, $0.030 cached input and $0.35 output, automatic prompt caching, no top-up fees, and published time to first token and throughput; for companies, self-optimizing inference that trains, tunes and optimizes models around a customer's workload, sold by contact.
What it does
- Model inference
- Model hosting
- Model training
Features 4
- Open models, low costRuns open models for lowest cost and latency behind an API.
- Change the endpointKeep existing code; add the Isoquant key and update the base URL.
- Automatic prompt cachingCached input tokens are billed at a lower published rate.
- Self-optimizing for companiesTrains, tunes and optimizes models around a customer's workload.
Sources
- Pricing
-
Official documentation · 18 Oct 2026
Home page: GLM-5.3-Flash at $0.09 per 1M input tokens, $0.030 per 1M cached input tokens and $0.35 per 1M output tokens, with no top-up fees; company optimization is by contact.
- Description
-
Official documentation · 18 Oct 2026
Home page: "Keep your code. Change the endpoint."; "Add your Isoquant key and update the base URL."
- What it does
-
Official documentation · 18 Oct 2026
Home page: "Keep your code. Change the endpoint."; "Add your Isoquant key and update the base URL."
- Constraints
-
Official documentation · 18 Oct 2026
Derived from the security facts on this page, from the same source.
- Security
-
Official documentation · 18 Oct 2026
Privacy policy: inputs, outputs and derived cache states are not used for model training without written consent; the home page lists SOC 2 as in progress, so no certification is recorded; a vendor claim.
- Sold within
-
Self-serve API key with published per-token prices; the company optimization offer is by contact.
- Name
-
Official documentation · 18 Oct 2026
Name as published on the product page.
- URL
-
Official documentation · 18 Oct 2026
Product page, read 2026-10-18.
- Deployment
-
Official documentation · 18 Oct 2026
The pages read state no deployment model beyond the vendor-run service; saas is recorded as the vendor-operated reading.
- Interfaces
-
Official documentation · 18 Oct 2026
Home page: a chat-completions API example at api.isoquant.ai and a Sign in link.
Open Isoquant in the directory
Something wrong here? Send a correction — quote this product id: isoquant-inference.