Coding subscriptions
From 49 €/month. Up to 250 tokens/s.
One API for the best open-weight coding models, upgraded as better ones ship. Drops into Claude Code, Codex or Cursor.
Not sure yet? See our Comparison and Pricing Plans
OpenCode 
Continue.dev 
Mistral Vibe 
Pi Included in coding plans
Frontier models
Explore the current open-weight models. Melious adds newer models to your plan as they become available.
Plan comparison
Compare coding plans
Explore monthly token volume and generation speed across Melious, Claude Max and Claude Ultra.
Estimated monthly token volume
- Claude Max About 250M tokens
- Melious Basic 700M to 7B tokens, depending on the model
- Claude Ultra About 1B tokens
- Melious Pro 2.5B to 25B tokens, depending on the model
Melious Unlimited 3x 3 concurrent generations Unlimited tokens
Token volumes are estimates and vary by model. Claude figures are public estimates.
European sovereignty
Your code stays in Europe.
Coding plans run on European infrastructure. The same data handling applies to agent traffic.
- EU/EEA only Infrastructure and jurisdiction
- No logging or training For prompts, code and completions
Monthly plans
Choose a coding plan
Pick a token budget or a plan based on parallel requests.
-
Token-based
Basic
99 €49 € / month700M–7B estimated tokens
-
Token-based
Pro
199 €149 € / month2.5B–25B estimated tokens
-
Concurrency-based
Unlimited 3x
699 €499 € / month3 parallel generations
-
Concurrency-based
Unlimited 6x
1.399 €999 € / month6 parallel generations
This offer is intended for businesses as defined by § 14 BGB (German Civil Code) · all prices plus statutory VAT. Token estimates vary by model.
Setup
Keep your coding tools.
Point your OpenAI- or Anthropic-compatible tool at Melious and use your Melious API key.
Read the integration docs (opens in a new tab).env
BASE_URL=https://api.melious.ai/v1
API_KEY=sk-mel-...
MODEL_ID=<MODEL_ID>Integrations
Up and running in three commands.
Claude Code
Anthropic's coding agent for the terminal, running on a Melious model.
Claude Code docs (opens in a new tab)What our customers say
We put Melious through a thorough hands-on test. With GLM 5.2 we measured more than 350 tokens per second, and a development task that used to take 45 minutes was done in about 7. It stayed stable throughout, entirely on European infrastructure, and our data is not used for training. That convinced us to integrate Melious as a provider in ALBERT | AI, the AI in agorum core pro.
Oliver Schulze CEO & CTO of agorum 
If inference stalls, the mule stumbles. Running agents smoothly only works when the inference keeps up. With Melious, we can be sure it does.
Bernd Paulus CTO & Co-Founder of mulios Melious isn’t just a supplier for us, they’re responsive, reliable, and always thinking along with us.
Robert Keus CEO of GreenPT FAQ
- Every coding plan includes selected frontier open-weight models built and tested for agentic coding. When a better open-weight coding model ships, it's added to your plan automatically, at no extra cost.
- No. Set your base URL to api.melious.ai/v1 and use your Melious API key — anything OpenAI- or Anthropic-compatible works unchanged, which covers 20+ tools including Claude Code, Cursor, Codex and Cline. Prefer a guided setup? The Melious CLI configures your tools for you.
- Token-based plans (Basic, Pro) bill by how much usage your requests draw each month — good if your usage varies day to day. Concurrency-based plans (Unlimited 3x/6x) instead cap how many requests can run in parallel and remove the token ceiling — good if you have rather consistent workloads and don't want to think about usage.
- No, never. Your prompts, code and completions stay on European infrastructure and are never logged or used for training — ours or anyone else's.
- Yes — nothing is different here. Coding plans run on the exact same European infrastructure and jurisdiction guarantee as the rest of Melious: EU/EEA only, no US legal exposure, no exceptions for agentic traffic.