---
license: apache-2.0
language:
- en
- fr
base_model: TinyLlama/TinyLlama-1.1B-Chat-v1.0
tags:
- esp32
- esp-idf
- embedded
- embedded-systems
- electronics
- microcontroller
- arduino
- firmware
- edge-ai
- fine-tuned
- lora
- code-generation
- mascarade
library_name: transformers
pipeline_tag: text-generation
---

# Mascarade ESP32

Fine-tuned **TinyLlama-1.1B-Chat** model specialized in **ESP32** microcontroller development.

Part of the [Mascarade](https://github.com/electron-rare/mascarade) ecosystem — an agentic LLM orchestration system with domain-specific fine-tuned models for embedded systems and electronics.

## Training details

| Parameter | Value |
|-----------|-------|
| Base model | `TinyLlama/TinyLlama-1.1B-Chat-v1.0` |
| Method | LoRA (PEFT) — merged into full weights |
| LoRA rank (r) | 16 |
| LoRA alpha | 32 |
| LoRA dropout | 0.05 |
| Target modules | q_proj, k_proj, v_proj, o_proj |
| Epochs | 2 |
| Training steps | 30 |
| Final train loss | 1.3873 |
| Dataset | ShareGPT format, domain-specific ESP32 examples |
| GPU | Quadro P2000 (5 GB VRAM) |
| Framework | Hugging Face Transformers + PEFT |

## Usage

```python
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("electron-rare/mascarade-esp32")
tokenizer = AutoTokenizer.from_pretrained("electron-rare/mascarade-esp32")

messages = [{"role": "user", "content": "How do I configure deep sleep on ESP32-S3?"}]
inputs = tokenizer.apply_chat_template(messages, return_tensors="pt")
outputs = model.generate(inputs, max_new_tokens=512)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```

## Related models

| Model | Domain | Base |
|-------|--------|------|
| [mascarade-iot](https://hf.co/electron-rare/mascarade-iot) | IoT general | Qwen2.5-Coder-1.5B |
| [mascarade-spice](https://hf.co/electron-rare/mascarade-spice) | SPICE circuit simulation | TinyLlama-1.1B |
| [mascarade-platformio](https://hf.co/electron-rare/mascarade-platformio) | PlatformIO development | TinyLlama-1.1B |

## Datasets

All training datasets are available under [clemsail on Hugging Face](https://hf.co/clemsail).

## 🇪🇺 EU AI Act transparency

This adapter is provided as a fine-tuned LoRA under the AI Act framework
(Regulation EU 2024/1689). Compliance metadata:

| Field | Value |
|---|---|
| Provider | L'Électron Rare (clemsail / electron-rare) |
| Role under AI Act | GPAI provider for this adapter |
| Base model | `TinyLlama/TinyLlama-1.1B-Chat-v1.0` — see upstream provenance |
| Adapter type | LoRA / PEFT — adapter weights only; base unchanged |
| Training data origin | L'Électron Rare proprietary technical corpus + curated public docs |
| License | Apache-2.0 (adapter). Upstream base licence applies separately. |
| Intended use | ESP32 / ESP-IDF firmware |
| Out of scope | Healthcare diagnosis, legal advice, autonomous safety-critical decisions, generation of malicious code |
| Risk classification | Limited risk — Article 50 transparency obligations apply |
| Copyright respect | Training data does not include scraped copyrighted material. Opt-out signals (robots.txt, ai.txt) are honoured for web-sourced data. |
| Full provenance | https://github.com/ailiance/ailiance/tree/main/docs/provenance |
| Contact | postmaster@saillant.cc — biased output reports, copyright concerns, etc. |

⚠️ **You are using an AI model.** Outputs may be inaccurate, biased or
fabricated. Do not act on them without independent verification, especially
in regulated domains.