---
license: apache-2.0
language:
  - en
tags:
  - tool-routing
  - nlm
  - mcp
  - orchestrator
  - nano-language-model
pipeline_tag: text-generation
model-index:
  - name: maaza-nlm-orchestrator-9.6m
    results:
      - task:
          type: text-generation
          name: Tool Routing
        metrics:
          - type: accuracy
            value: 70
            name: Tool Selection Accuracy
          - type: latency
            value: 70
            name: Average Latency (ms)
---

# maaza-nlm-orchestrator-9.6m

**70% tool-routing accuracy · 70ms average latency · 9.60M parameters**

The fastest production-ready orchestrator under 20M parameters. The official routing brain for the MCPBodega ecosystem.

---

## NLM Taxonomy (CycleCore, 2025)

| Category | Parameters | Typical Capability |
|----------|------------|--------------------|
| **NLM** | <10M | Routing, classification, orchestration |
| MLM | 10–250M | Structured extraction |
| SLM | 250M–1.5B | Reliable reasoning + extraction |
| LLM | >1.5B | General-purpose reasoning |

maaza-nlm-orchestrator-9.6m is the current flagship of the NLM category.

> For the full taxonomy and research, see the [Maaza Paper v0.7](https://cyclecore.ai/papers/MAAZA_PAPER_v0.7_dark.pdf)

---

## Model Card

| Metric | Value |
|--------|-------|
| Parameters | 9,600,000 |
| Architecture | 7-layer Transformer decoder, SwiGLU, RoPE |
| Hidden size / Heads | 320 / 8 |
| Vocabulary | 8,000 (BPE, tool-aware) |
| Context length | 512 tokens |

Trained exclusively on 36 real, production-ready MCP tools from MCPBodega (Doom, Puppeteer, code execution, file I/O, database queries, etc.). No synthetic or placeholder tools.

---

## Benchmarks (verified December 2025)

| Metric | Value | Notes |
|--------|-------|-------|
| Tool selection accuracy | **70%** | Held-out evaluation (100 samples) |
| Valid structured output | **88%** | Parsable JSON output |
| Average inference latency | **70ms** | RTX 4080, fp16, batch=1 |
| Parameters | 9,600,000 | |

> Initial reports of 90% were aspirational targets. Verified held-out performance is 70% tool selection with 88% valid output.

### Comparison

| Model | Parameters | Tool Accuracy | Latency |
|-------|------------|---------------|---------|
| **maaza-nlm-orchestrator-9.6m** | **9.6M** | **70%** | **70ms** |
| NVIDIA Orchestrator-8B | 8B | 78% | ≥800ms |
| Gorilla-7B | 7B | 52–58% | 1–3s |
| ToolLlama-7B | 7B | 48–55% | 2–4s |

Ranks **#1 under 20M parameters** on latency-adjusted tool routing.

---

## One-line deployment

```bash
mcpbodega deploy nano-orchestrator
```

---

## Usage Example (PyTorch)

```python
from model import MaazaNanoModel, MaazaNanoConfig
from tokenizer import BPETokenizer
import torch, json

tokenizer = BPETokenizer.load("tokenizer.json")
config = MaazaNanoConfig(**json.load(open("config.json")))
model = MaazaNanoModel(config)
model.load_state_dict(torch.load("model.pt", weights_only=True))
model.eval().cuda()

prompt = "<|user|>search for cats on the internet<|assistant|>"
input_ids = torch.tensor([tokenizer.encode(prompt)]).cuda()

with torch.no_grad():
    for _ in range(64):
        logits = model(input_ids)["logits"]
        next_token = logits[0, -1].argmax(-1)
        input_ids = torch.cat([input_ids, next_token[None, None]], dim=-1)
        if next_token.item() in tokenizer.special_tokens.values():
            break

print(tokenizer.decode(input_ids[0].tolist()))
# → [{"tool": "web_search", "params": {"query": "cats"}}]
```

---

## Supported Tools (36)

| Tool | Description |
|------|-------------|
| `web_search` | Search the web |
| `web_fetch` | Fetch URL content |
| `file_read` | Read local files |
| `file_write` | Write local files |
| `code_execute_python` | Run Python code |
| `code_execute_bash` | Run shell commands |
| `code_execute_js` | Run JavaScript |
| `email_send` | Send emails |
| `slack_send` | Send Slack messages |
| `calendar_add` | Create calendar events |
| `database_query` | Query databases |
| `puppeteer_navigate` | Browser navigation |
| `puppeteer_click` | Browser clicks |
| `puppeteer_screenshot` | Take screenshots |
| `doom_mcp` | Play Doom |
| `bitchat_send` | BLE mesh chat |
| `voice_mcp` | Text-to-speech |
| `maaza_extract_json` | Extract structured data |
| `json_validate` | Validate JSON |
| `csv_parse` | Parse CSV files |
| `regex_match` | Pattern matching |
| `calculator` | Math operations |
| `weather_lookup` | Weather data |
| `crypto_lookup` | Crypto prices |
| `stock_lookup` | Stock prices |
| `news_fetch` | News headlines |
| `mcpbodega_chat` | MCPBodega chat rooms |
| `mcpbodega_deploy` | Deploy MCPs |
| `mcpbodega_list` | List MCPs |
| `github_issue` | Create GitHub issues |
| `scratchpad_mcp` | Temporary storage |
| `health_check` | Service health checks |
| `cyclecore_terminal` | Terminal commands |
| `image_caption` | Image descriptions |
| `slmbench_query` | Benchmark queries |
| `translator` | Translation |

---

## License

Apache 2.0

---

## Citation

```bibtex
@misc{cyclecore2025maaza-nlm,
  author       = {CycleCore},
  title        = {Maaza NLM Orchestrator 9.6M: 70% Tool-Routing at 70ms},
  year         = {2025},
  publisher    = {Hugging Face},
  url          = {https://huggingface.co/cyclecore/maaza-nlm-orchestrator-9.6m}
}
```

---

Built with MCPBodega · December 2025