We’re thrilled to announce a strategic partnership with Opper AI, the leading European AI gateway and control plane for agents. Novita AI’s inference catalog is now available on Opper, giving the Opper developer community direct access to 80+ Novita open-weight models (such as DeepSeek V4, Kimi K2.6, GLM-5, MiniMax M3, Llama 4, and Qwen) through a single, unified API.
What This Partnership Brings
The European AI gateway built for agents. Opper AI was founded in Stockholm, Sweden, and is purpose-built for teams running AI agents in production. At its core, Opper is one API in front of 300+ leading AI models, hosted in the EU and designed from the ground up with GDPR compliance built in. When your team needs access to the latest frontier and open-weight models without standing up your own proxy infrastructure, Opper is where European builders go first.
Full AI control plane, not just a router. What makes Opper distinctive is the layer above the routing: complete observability into every call, token, and session; real-time PII masking and content filtering; intelligent routing across models; automatic fallbacks when a provider hits rate limits or outages; budget caps; and full audit trails. For teams operating agents in regulated environments or at scale, that means one integration covers routing, reliability, and governance—no additional tooling required.
EU compliance handled by default. Opper provides a single data processing agreement that covers every model in the catalog. EU data residency is configurable at the route level, and every request can be pinned to EU-hosted infrastructure where needed. For European teams managing GDPR review cycles, having one subprocessor instead of one per model is a meaningful simplification.
Drop-in compatible with the SDKs you already use. Opper works with the OpenAI, Anthropic, and Google AI SDKs without code changes. One API key and the model string are all you need to start routing calls through Opper—including to Novita-hosted models.
Novita AI joins Opper’s gateway as one of its broadest single-provider catalogs, covering models from close to twenty different labs. Developers building on Opper can now reach Novita-hosted models with the same single API key they already use—no separate Novita account, no additional integration, just a model string change.
Novita’s contribution to the Opper catalog is open-weight breadth and speed. We serve frontier reasoning models, long-context models with 1M-token windows, coding specialists, multimodal models, and specialist checkpoints including OCR models. We make a point of getting new releases production-ready fast: when a new open model lands from DeepSeek, Moonshot, Z.ai, or MiniMax, Novita is consistently among the first providers to have it callable via a stable, production-grade API.
“We’re glad to welcome Novita to Opper. They bring one of the broadest open-weight catalogs anywhere—from DeepSeek and Kimi to GLM, MiniMax, and Llama—and they’re consistently among the first to get new releases production-ready. That gives our developers a deep bench of models they can reach the moment they need them.”
— Göran Sandahl, Co-founder and CEO, Opper AI
“Developers come to Novita because new open models show up fast and just work in production. Opper puts that catalog in front of a large community of builders through one API, with routing, observability, and governance handled, so teams can adopt new models the day they land.”
— Junyu Huang, COO, Novita AI
Models Available on Opper Today
All Novita routes on Opper are US-hosted. Pricing is in USD per 1M tokens and subject to change.
| Model | Context | Input / 1M | Output / 1M |
|---|---|---|---|
| novita/deepseek-v4-flash | 1.0M | $0.14 | $0.28 |
| novita/deepseek-v4-pro | 1.0M | $1.74 | $3.48 |
| novita/deepseek/deepseek-r1-0528 | 164K | $0.70 | $2.50 |
| novita/deepseek-v3.2 | 164K | $0.27 | $0.40 |
| novita/glm-5 | 205K | $1.00 | $3.20 |
| novita/glm-4.7 | 205K | $0.60 | $2.20 |
| novita/glm-4.7-flash | 200K | $0.07 | $0.40 |
| novita/kimi-k2.6 | 262K | $0.95 | $4.00 |
| novita/moonshotai/kimi-k2.7-code | 262K | $0.95 | $4.00 |
| novita/minimax/minimax-m3 | 1.0M | $0.30 | $1.20 |
| novita/minimax/minimax-m2.7 | 205K | $0.30 | $1.20 |
| novita/meta-llama/llama-4-maverick-17b-128e-instruct-fp8 | 1.0M | $0.27 | $0.85 |
| novita/meta-llama/llama-4-scout-17b-16e-instruct | 131K | $0.18 | $0.59 |
| novita/meta-llama/llama-3.3-70b-instruct | 131K | $0.14 | $0.40 |
| novita/qwen/qwen-2.5-72b-instruct | 32K | $0.38 | $0.40 |
| novita/google/gemma-4-31b-it | 262K | $0.14 | $0.40 |
| novita/openai/gpt-oss-120b | 131K | $0.05 | $0.25 |
| novita/openai/gpt-oss-20b | 131K | $0.04 | $0.15 |
| novita/baidu/ernie-4.5-300b-a47b-paddle | 123K | $0.28 | $1.10 |
| novita/nvidia/nemotron-3-nano-30b-a3b | 262K | $0.05 | $0.20 |
| novita/deepseek-ocr-2 | 128K | $0.03 | $0.03 |
For the full catalog, see the Novita provider listing on Opper.
Get Started
If you’re already on Opper, Novita models are available today—just use a novita/ model string. For new users, Opper is drop-in compatible with the OpenAI SDK:
import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.OPPER_API_KEY,
baseURL: "https://api.opper.ai/v3/compat",
});
const completion = await client.chat.completions.create({
model: "novita/deepseek-v4-flash",
messages: [{ role: "user", content: "Hello" }],
});
console.log(completion.choices[0].message.content);See Opper’s quick start for routing, fallbacks, and evaluations. For Novita’s full model catalog and direct API access, visit novita.ai/models.
Start building with Novita on Opper →
Opper AI is the European AI gateway and control plane for agents—one EU-hosted, GDPR-compliant API across 300+ models, with smart routing, automatic fallbacks, built-in observability and evaluations, real-time guardrails, and full OpenAI SDK compatibility. Founded in Stockholm, Sweden.
Novita AI is an AI cloud platform headquartered in San Francisco, serving 200+ open-weight and multimodal models across text, image, audio, and video through an OpenAI-compatible API, alongside dedicated and serverless GPU infrastructure and agent sandboxes. Novita is SOC 2 Type II audited and does not train on customer data.
