| --- |
| license: apache-2.0 |
| language: |
| - en |
| tags: |
| - ngari |
| - sovereign-ai |
| - edge-ai |
| - qwen2.5 |
| - lora |
| base_model: Qwen/Qwen2.5-1.5B-Instruct |
| --- |
| |
| # NGARi Tool — Tool-Calling 1.5B |
|
|
| 1.5B tool-calling model: Qwen2.5-1.5B-Instruct + tool-format LoRA. 100% tool detection / name / params validity on the NGARi tool-format eval (20 examples). Serves as the NGARi agent tool-mode model in production. |
|
|
| ## Provenance (verified Aug 3, 2026) |
|
|
| | Attribute | Value | |
| |-----------|-------| |
| | Base model | `Qwen/Qwen2.5-1.5B-Instruct` (Apache 2.0) — pinned in `adapter_config.json` | |
| | LoRA | rank 32, alpha 64, dropout 0.05, all linear projections | |
| | Synthetic data teacher | `qwen3:8b` | |
| | License | Apache 2.0 (NGARi-authored artifacts) | |
| | Hardware validated | aarch64 / NVIDIA Jetson AGX Orin, 8GB RAM, air-gap verified | |
|
|
| > Note: Google Gemma models were **served only** on NGARi hardware and were never used in NGARi training. All training used the Apache-2.0 Qwen2.5 lineage. |
|
|
| ## Evaluation |
|
|
| ### ngari-tool-stable_tool_eval.json |
|
|
| ```json |
| { |
| "model": "ngari-tool:stable", |
| "num_examples": 20, |
| "total_examples": 20, |
| "tool_detection_rate": 1.0, |
| "tool_name_accuracy": 1.0, |
| "params_validity_rate": 1.0, |
| "tool_detected": 20, |
| "name_correct": 20, |
| "params_valid": 20, |
| "avg_latency_sec": 3.17, |
| "total_time_sec": 63.4 |
| } |
| ``` |
|
|
| ## Files |
|
|
| | File | Purpose | |
| |------|---------| |
| | `model-*.safetensors` (+ config) | Merged full model — use with Transformers | |
| | `adapter_model.safetensors` | PEFT LoRA adapter — apply on the base | |
| | `*-q4_K_M.gguf` / `*-f16.gguf` | GGUF — use with Ollama / llama.cpp | |
|
|
| ## Usage |
|
|
| ```bash |
| # Ollama (GGUF) |
| ollama create ngariai/ngari-tool:q4_K_M -f Modelfile |
| |
| # Transformers (merged) |
| from transformers import AutoModelForCausalLM, AutoTokenizer |
| model = AutoModelForCausalLM.from_pretrained("ngariai/ngari-tool") |
| |
| # PEFT adapter (apply on base) |
| from peft import PeftModel |
| base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen2.5-1.5B-Instruct") |
| adapter = PeftModel.from_pretrained(base, "ngariai/ngari-tool") |
| ``` |
|
|
| ## Sovereign AI |
|
|
| Trained and verified on user-owned edge hardware with zero cloud dependency. Part of the NGARi Sovereign Business Operating System (NS-BOS) — see https://github.com/ngariai/ns-bos-kernel for the Apache 2.0 kernel. |
|
|
|
|
