Who makes Inference.net?

Inference.net is made by Inference.net.

Does Inference.net work in the EU?

Unverified. No explicit evidence found. The pages show no supported-countries list, EU exclusion or EU data residency information. Last verified 14 Nov 2026.

How much does Inference.net cost?

API pricing. Pay per token at pass-through provider prices (e.g. GLM-5.2 $1.30 in / $4.40 out per 1M tokens). Plans: $9/mo with $10 Gateway credit; $49/mo with $60 credit; pay as you go after credit. Licence: Proprietary.

Where was Inference.net shown?

In 1 creator video by Matt Wolfe. Most recently by Matt Wolfe on 1 Jul 2026, at 25:10 in "GLM-5.2: The Complete Guide to the Best Open-Source Model".

Inference.net

by Inference.net

Try it Website

OpenAI-compatible inference API and gateway for open-weight and closed models, with provider failover and pass-through pricing.

UnverifiedPlatforms & InfraAPI

Updated 14 Nov 2026EU status checked 14 Nov 2026Last shown 1 Jul 2026

Inference.net at a glance

What is Inference.net?: Inference.net is an api from Inference.net. OpenAI-compatible inference API and gateway for open-weight and closed models, with provider failover and pass-through pricing.
Who makes Inference.net?: Inference.net is made by Inference.net.
Does Inference.net work in the EU?: Unverified. No explicit evidence found. The pages show no supported-countries list, EU exclusion or EU data residency information. Last verified 14 Nov 2026.
How much does Inference.net cost?: API pricing. Pay per token at pass-through provider prices (e.g. GLM-5.2 $1.30 in / $4.40 out per 1M tokens). Plans: $9/mo with $10 Gateway credit; $49/mo with $60 credit; pay as you go after credit. Licence: Proprietary.
Where was Inference.net shown?: In 1 creator video by Matt Wolfe. Most recently by Matt Wolfe on 1 Jul 2026, at 25:10 in “GLM-5.2: The Complete Guide to the Best Open-Source Model”.

Sources: the vendor's own pages (linked above) and creator videos at the linked timestamps. EU status is information, not legal advice; see how we check.

About

Inference.net is an inference provider built around open-weight models, offering a single API for models such as Kimi K3, GLM-5.2 and DeepSeek V4 Flash. It also lists closed models, including Claude Opus 5 Fast and GPT-5.6 Sol Fast. The endpoints are OpenAI-compatible, so they can be used from applications and coding agents, including Claude Code and Codex.

The service works as a gateway. One API key covers all supported providers, and each model routes across providers and moves to the next healthy route if one slows or fails. The site states that provider pricing is passed through with no per-request markup. Listed per-million-token prices include GLM-5.2 at $1.30 input / $4.40 output and Kimi K3 at $3.00 / $15.00. A YouTube creator mentioned it as a way to mirror production traffic to GLM-5.2 before switching models.

On data handling, the site says prompts and outputs are not used to train models, Zero Data Retention is the default (prompts and outputs are not retained after each request), and keys are scoped to a single project with one-click revocation. It offers usage metering. Subscription plans include monthly Gateway credit, with pay-as-you-go after the credit is used. The page text does not give details on documentation, regions or data residency.

Key features