🇵🇱 Made in Poland · Powered by Ollama · Self-hosted

Language & Code
Reimagined

Three specialized AI models engineered for real-world production use. Mika v3 handles language. Core Code v1 writes your software. TaskMind v1 orchestrates your agents. All self-hosted, all private, all Polish-built.

Start for free Sign in to dashboard
3
Specialized models
100%
Self-hosted, private
API
OpenAI compatible
No vendor lock-in
Mika v3 · Language Support
Core Code v1 · Code Generation
REST API · Ollama Compatible
OpenAI Format · Drop-in Replacement
Self-hosted · Your Data Stays Private
Token Budgets · Per-user Quotas
Dark & Light Theme
Multi-language · EN / PL
Mika v3 · Language Support
Core Code v1 · Code Generation
REST API · Ollama Compatible
OpenAI Format · Drop-in Replacement
Self-hosted · Your Data Stays Private
Token Budgets · Per-user Quotas
Dark & Light Theme
Multi-language · EN / PL
Our Models

Three minds.
One platform.

Three purpose-built AI models engineered for distinct professional domains — not interchangeable generalists, but precision-tuned specialists operating on a shared, fully self-hosted inference infrastructure. Built in Poland. 🇵🇱

Language Model
Mika v3
coreinfra/mika · v3.0 stable

Mika v3 is your intelligent language partner — trained to communicate with clarity, empathy and precision in both Polish and English. Whether you're drafting professional correspondence, translating technical documents, summarizing reports, or just need a thoughtful sounding board, Mika delivers consistently human-quality text that sounds like you, not a machine. Mika employs a proprietary live continuous learning mechanism and on-the-fly training innovation, supervised by its non-public twin model Ełe — a dedicated fine-tuning supervisor that monitors inference quality, detects distribution drift, and triggers targeted weight updates without service interruption.

  • Bilingual Polish–English with native-level fluency and idiomatic accuracy
  • Tone adaptation — formal, casual, technical, persuasive, empathetic
  • Document drafting: emails, reports, proposals, summaries, social posts
  • Real-time grammar & style correction with explanations
  • Context-aware memory across long conversations
  • Low hallucination rate — Mika says "I don't know" when she doesn't
  • Continuous live learning supervised by twin model Ełe (non-public) — self-improving without downtime
128K
Context window
~0.7
Temperature
PL/EN
Languages
Code Model
Core Code v1
coreinfra/core-code · v1.0 stable

Core Code v1 is a developer-grade AI model tuned specifically for software engineering tasks. It understands architecture, not just syntax. Give it a task description and it returns production-ready code with proper error handling, tests, and documentation — across every major language and framework in use today. Like all MikaAI models, Core Code v1 incorporates an automated live-learning pipeline — real-world inference patterns are continuously analysed and fed back into incremental training cycles supervised by the twin model Ełe, keeping the model sharp against evolving codebases and emerging frameworks.

  • Full-stack code generation: frontend, backend, APIs, database schemas
  • Intelligent debugging — explains the root cause, not just the fix
  • Code review & refactoring with clean-code principles (SOLID, DRY)
  • Architecture consulting: chooses the right pattern for the problem
  • Test generation: unit, integration & edge cases automatically
  • Understands PHP, Python, TypeScript, Go, Rust, SQL and 20+ more
128K
Context window
~0.2
Temperature
20+
Languages
Agent Model
TaskMind v1
coreinfra/taskmind · v1.0 stable

TaskMind v1 is purpose-built for AI agent frameworks — designed to understand tool calls, multi-step instructions, structured JSON output, and long-horizon task planning. Whether you're running Hermes, OpenClaw, or any other orchestration layer, TaskMind integrates as a drop-in model endpoint that speaks the same language as your agents. TaskMind's tool-calling accuracy benefits directly from the shared live-learning infrastructure: the twin model Ełe continuously evaluates JSON schema compliance across all inference calls, triggering micro-fine-tuning passes that reinforce structured output fidelity over time.

128K
Context window
~0.1
Temperature
JSON
Native output
ReAct
Agent pattern
  • Native function/tool calling — returns valid JSON schemas on first try
  • Chain-of-thought reasoning for complex multi-step task decomposition
  • Compatible with Hermes, OpenClaw, AutoGen, CrewAI and LangChain agents
  • Persistent context handling — maintains state across long agent sessions
  • Low latency structured output — optimised for high-frequency agent loops
  • Supports ReAct, Reflexion and Plan-and-Execute agent patterns out of the box
AI Agent Ecosystem

TaskMind v1 plugs directly into your existing agent infrastructure. Download the official integrations for the most popular frameworks.

Download integrations
How It Works

Up and running
in minutes.

STEP 01
👤
Create your account
Register for free and receive 1,000 tokens per day instantly — no credit card required. Start chatting with Mika v3 within seconds.
STEP 02
Choose your model
Select Mika v3 for language tasks or Core Code v1 for development work. Switch models mid-conversation as your needs change.
STEP 03
🔌
Connect via API
Generate your API token and integrate into your apps, scripts or IDE. Fully compatible with OpenAI and Ollama client libraries.
Developer API

Drop-in
replacement.

If you already use OpenAI or Ollama, switching to MikaAI takes one line of code. Change the base URL and your API token — everything else stays the same. Full streaming support included.

  • endpoint POST /api/chat — Ollama format, streaming
  • endpoint POST /api/chat/completions — OpenAI format
  • endpoint GET /api/tags — list available models
  • endpoint POST /api/generate — single-turn generation
  • endpoint GET /api/stats — usage metrics
View full API docs →
api_example.py
# MikaAI — drop-in OpenAI replacement
from openai import OpenAI

client = OpenAI(
  base_url="https://ai.coreinfra.eu/api",
  api_key="your-token-here",
)

response = client.chat.completions.create(
  model="mika-v3",
  messages=[{
    "role": "user",
    "content": "Review this code and explain it",
  }],
  stream=True,
)

for chunk in response:
  print(chunk.choices[0].delta.content)
Platform Features

Everything you
actually need.

🔒
Private by design
Your data never leaves your server. No telemetry, no training on your conversations. Complete data sovereignty.
📊
Token budgets
Per-user daily limits with real-time tracking. Upgrade to paid plans for higher quotas. Admins get unlimited access.
Rate limiting
OpenAI-compatible RPM, TPM and RPD limits. Configurable from the admin panel. Fair usage for all users.
🌐
REST API
Fully documented Ollama & OpenAI compatible API. Integrate into any app, script or IDE in minutes.
🌗
Dark & light theme
Beautiful dark and light modes that sync with your system preference or can be set manually per account.
🇵🇱
Multilingual UI
Full Polish and English interface. Language preference saved per account. Admin sets the default.

Ready to start?
It's free.

1,000 tokens every day, zero credit card required. Upgrade when you need more.

Create free account Already have an account