Opper AI partners with Inceptron for price-performance on EU-hosted inference

Stockholm, Sweden, September 2026. Opper AI and Inceptron are partnering to bring Inceptron's EU-hosted inference to the Opper AI gateway. Builders can now reach Inceptron's catalogue through the Opper API, from GLM-5.3 and GLM-5.3-Flash to GLM-5.2, DeepSeek V4 Flash, Kimi K2.6 and the coding-tuned Kimi K2.7 Code.

Why this matters

Frontier open weights at the low end of the European field. The roster is curated rather than long-tail: both halves of the GLM-5.3 pair, GLM-5.2 and DeepSeek V4 Flash at a million tokens of context, and Moonshot's Kimi K2.6 alongside Kimi K2.7 Code for agent and coding loops. Price-performance is the pitch, and for most of what it serves Inceptron comes in at the cheapest European route on the gateway. Rosters and prices both move, so the table further down is fetched live and is the place to check rather than take our word for it.

Where the price comes from. Inceptron came at inference from the optimisation side rather than the fleet side: batched serving tuned to hold p95 under load, quantized variants, and kernels auto-tuned against the hardware a model actually runs on, with an optimisation compiler in early access on top. Teams that outgrow the serverless endpoints can bring their own checkpoints or move to dedicated GPUs on the same API.

Clear data handling. Inceptron AB is a Swedish processor, and on an EU configuration its DPA commits that request and response payloads are processed inside the EU/EEA and not transferred out. Payloads are not stored after processing unless you switch retention on yourself, and customer content is never used to train models. Full residency and retention detail per route lives on the Inceptron provider page.

More than a router: the AI control plane. Routing to Inceptron is just the entry point. Every call through Opper runs on its AI control plane: intelligent routing across models and regions, full observability, real-time PII masking and content filtering, budget caps, and audit trails. Pin Inceptron for a task or set it as a fallback, and because Inceptron is OpenAI-compatible, getting there is a model string, not a migration.

"Inceptron has been working on the hard part of inference, making a given model run well on given hardware, since before most of this catalogue existed. What that buys our developers is frontier open weights inside the EU at prices that hold up against anything in Europe, which is exactly the combination teams keep asking us for."

— Göran Sandahl, Co-founder and CEO, Opper AI

"We built Inceptron so teams get the best price-performance out of open models without running the infrastructure or the optimisation themselves. Opper puts that in front of a wide community of builders through a single API, and we are looking forward to what they build on it."

— Lucas Ferreira, Co-founder and CEO, Inceptron

Models live today

The catalog below is fetched live from Opper's model API and filtered to Inceptron-hosted models. Availability, context windows, and pricing stay in sync with what's actually callable through Opper. All Inceptron routes on Opper are EU-hosted.

Get started

Paste this into your coding agent (Claude Code, Cursor, Codex, and more) and it will set up Opper and route to Inceptron for you:

Use curl to download, read and follow: https://skills.opper.ai
Then set up Opper to use Inceptron as the provider, e.g. inceptron/glm-5.2.

Prefer a direct call? Opper is drop-in compatible with the OpenAI, Anthropic, and Google SDKs, so one API key and the model string are all you need:

OpenAIAnthropicGeminicURL

1import OpenAI from "openai";
2
3const client = new OpenAI({
4  apiKey: process.env.OPPER_API_KEY,
5  baseURL: "https://api.opper.ai/v3/compat",
6});
7
8const completion = await client.chat.completions.create({
9  model: "inceptron/glm-5.2",
10  messages: [{ role: "user", content: "Hello" }],
11});
12console.log(completion.choices[0].message.content);

Follow the quick start in our docs for evaluations, fallbacks, and structured output.

About Inceptron

Inceptron (Inceptron AB) is a Swedish AI infrastructure company founded in Lund in 2021. It serves open-weight models for price-performance through an OpenAI-compatible API, on pay-per-token serverless endpoints and dedicated single-tenant GPU deployments, with inference inside the EU/EEA.

About Opper AI

Opper AI is the European AI gateway and control plane for agents: one EU-hosted, GDPR-compliant API across 700+ models, with smart routing, automatic fallbacks, built-in evaluations and observability, real-time guardrails, and full OpenAI SDK compatibility.