File size: 2,920 Bytes
b82520a 794ab9e b82520a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 |
---
license: apache-2.0
language:
- en
tags:
- code
- assistant
- ollama
- openai-compatible
- streaming
- voice
pipeline_tag: text-generation
inference: false
---
# GeniusPro Coder v1
**GeniusPro Coder v1** is a coding-focused AI assistant model built for providing intelligent code generation, explanation, and general-purpose AI assistance.
## Highlights
- Code generation, debugging, and explanation across multiple languages
- Natural conversational ability for non-code tasks
- OpenAI-compatible API (drop-in replacement for existing tooling)
- Streaming support for real-time token delivery
- Voice mode with concise, spoken-friendly responses
- Runs locally on consumer hardware via [Ollama](https://ollama.com)
## Intended Use
GeniusPro Coder v1 is designed for:
- **Code assistance** — generating, reviewing, debugging, and explaining code
- **Chat** — general-purpose question answering and conversation
- **Voice interaction** — concise, natural-language responses optimized for text-to-speech
It powers the GeniusPro platform, which includes a web-based chat dashboard and a real-time voice assistant.
### Supported Parameters
| Parameter | Description |
|-----------|-------------|
| `temperature` | Controls randomness (0.0 = deterministic, 1.0 = creative) |
| `top_p` | Nucleus sampling threshold |
| `max_tokens` | Maximum tokens to generate |
| `stop` | Stop sequences |
| `stream` | Enable streaming responses (SSE) |
### Available Endpoints
| Endpoint | Method | Description |
|----------|--------|-------------|
| `/v1/models` | GET | List available models |
| `/v1/chat/completions` | POST | Chat completions (streaming + non-streaming) |
| `/v1/voice` | WebSocket | Real-time voice interaction |
| `/health` | GET | Health check (no auth required) |
## Running Locally with Ollama
```bash
# Pull the model
ollama pull geniuspro-coder-v1
# Run interactively
ollama run geniuspro-coder-v1
# Serve via API
ollama serve
```
Once running, the model is available at `http://localhost:11434` with the same OpenAI-compatible API format.
## Infrastructure
GeniusPro Coder v1 runs on dedicated hardware for low-latency inference:
- **GPU**: NVIDIA RTX 5090 (32 GB VRAM)
- **Runtime**: [Ollama](https://ollama.com) for model serving
- **Gateway**: FastAPI reverse proxy with auth, rate limiting, and usage tracking
- **Deployment**: Ubuntu Server behind Nginx + Cloudflare Tunnel
## Limitations
- Optimized for English. Other languages may work but are not officially supported.
- Code generation quality varies by language — strongest in Python, JavaScript/TypeScript, and common web technologies.
- Not suitable for safety-critical applications without human review.
- Context window and output length are bounded by the underlying architecture.
## License
This model is released under the [Apache 2.0 License](https://www.apache.org/licenses/LICENSE-2.0).
|