// integrations

Drop-in.
Your stack doesn't change.

Helmcode is compatible with the OpenAI API. Change the base URL and key, and every OpenAI-compatible SDK, editor and framework you already use just works, now on private European infrastructure.

read_the_docs

// the change

Two lines. That's the migration.

No rewrite, no new SDK, no abstraction layer. Repoint the base URL and key, everything else in your codebase stays exactly as it is.

- base_url = "https://api.openai.com/v1"
+ base_url = "https://api.helmcode.com/v1"
- api_key  = OPENAI_API_KEY
+ api_key  = HELMCODE_API_KEY
  # model, messages, streaming, tools — all unchanged

// quickstart

A request, in your language.

The same call you already write against OpenAI, only the endpoint and the model id are ours.

from openai import OpenAI

client = OpenAI(
    base_url="https://api.helmcode.com/v1",
    api_key="sk-your-key-here",
)

resp = client.chat.completions.create(
    model="deepseek-v4-flash",
    messages=[{"role": "user", "content": "Hello from the EU"}],
)
print(resp.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.helmcode.com/v1",
  apiKey: process.env.HELMCODE_API_KEY,
});

const resp = await client.chat.completions.create({
  model: "deepseek-v4-flash",
  messages: [{ role: "user", content: "Hello from the EU" }],
});
console.log(resp.choices[0].message.content);

curl https://api.helmcode.com/v1/chat/completions \
  -H "Authorization: Bearer sk-your-key-here" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4-flash",
    "messages": [{"role": "user", "content": "Hello from the EU"}]
  }'

// works with

The tools you already run.

If it uses the OpenAI API, it can use Helmcode. A few of the editors, frameworks and SDKs teams point at us every day.

Editors & copilots

Cursor AI code editor
Zed Collaborative editor
OpenCode Terminal coding agent
Continue IDE autocomplete
Cline Autonomous coding

Frameworks

LangChain Agent orchestration
LlamaIndex RAG framework
Vercel AI SDK Streaming UI
Haystack Search & RAG
Pydantic AI Typed agents

SDKs & HTTP

OpenAI · Python Official SDK
OpenAI · JS/TS Official SDK
httpx · requests Python HTTP
Go · net/http Standard library
curl Anything else

…and any other OpenAI-compatible tool, there's no Helmcode SDK to learn.

// api surface

One endpoint per capability.

The full OpenAI-compatible surface at https://api.helmcode.com/v1, text, retrieval and speech, all behind the same key.

POST /v1/chat/completions Chat, streaming, tool & function calling

POST /v1/completions Legacy text completion

POST /v1/embeddings qwen3-embedding · 4096-dim vectors

POST /v1/rerank Cross-lingual semantic reranking

POST /v1/audio/speech Kokoro text-to-speech

POST /v1/audio/transcriptions Whisper speech-to-text

GET /v1/models List available models

// integrations faq

Integrations, answered.

What developers check before pointing their stack at Helmcode.

How much of my code has to change?

Two lines: the base URL and the API key. Helmcode implements the OpenAI API, so any OpenAI-compatible client, SDK or tool works unchanged, same request shape, same response shape, same streaming.

Do streaming and tool calling work?

Yes. Token streaming over SSE and native function/tool calling with the OpenAI JSON schema are supported on all chat models. structured outputs via response_format are supported too.

Which models can I call?

Any of the eight, by id, deepseek-v4-flash, qwen3.6, gemma4 for text, qwen3-embedding and rerank for retrieval, kokoro and whisper-large-v3 for speech. Just change the model field.

Can I point an existing OpenAI project at Helmcode?

Yes, that is the design. Set OPENAI_BASE_URL to https://api.helmcode.com/v1 and OPENAI_API_KEY to your Helmcode key, and your project runs on private EU infrastructure with no further changes.

Is there an SDK I have to install?

No Helmcode-specific SDK. Use the OpenAI SDK you already have, or any HTTP client. We keep to the standard so there is nothing new to learn.

// get started

START BURNING TOKENS

Skip the AI infra work. Deploy your first private inference endpoint today.

Flat rate. EU data. OpenAI API compatible.

book_a_call

Drop-in.Your stack doesn't change.