---
license: apache-2.0
language:
- en
library_name: transformers
pipeline_tag: text-generation
tags:
- qwen2.5
- sakthai
- plus
- tool-calling
- conversational
- function-calling
- merged
- rslor
- house-of-sak
- family
datasets:
- Nanthasit/sakthai-combined-v7
- Nanthasit/sakthai-combined-v8
base_model: Qwen/Qwen2.5-1.5B-Instruct
widget:
- text: What's the weather in Tokyo?
  output:
    text: '<tool_call>{"name": "get_weather", "arguments": {"location": "Tokyo"}}</tool_call>'
model-index:
- name: sakthai-plus-1.5b
  results:
  - task:
      type: text-generation
      name: Tool-Calling
    dataset:
      name: SakThai Bench v2 (500 rows, scorer multiset-selection-v2)
      type: Nanthasit/sakthai-bench-v2
    metrics:
    - type: selection
      value: pending
      name: Selection Accuracy
    - type: degenerate
      value: 0
      name: Degenerate Rate
---

# SakThai Plus 1.5B

**Next-generation tool-calling model — rsLoRA + all 7 module targets.** Built on Qwen2.5-1.5B-Instruct with improved training data and deeper fine-tuning.

Part of the [SakThai Model Family](https://huggingface.co/collections/Nanthasit/sakthai-model-family-6a64745450b12d421c1f9f02).

## 🚀 Improvements over v1 (sakthai-context-1.5b-merged)

| Feature | v1 (context-1.5b-merged) | Plus (this model) |
|---------|:------------------------:|:------------------:|
| LoRA method | Standard LoRA | **rsLoRA** (better rank scaling) |
| Linear targets | 4 modules | **All 7** (q, k, v, o, gate, up, down) |
| Dropout | 0.1 | **0.05** (lower, better retention) |
| Training data | v7 only (2,003 rows) | **v7 + v8** (2,962 rows) |
| Format | Tool XML | Tool XML (same, improved coverage) |

## Quick Start

```python
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained(
    "Nanthasit/sakthai-plus-1.5b",
    torch_dtype="auto",
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained("Nanthasit/sakthai-plus-1.5b")

messages = [{"role": "user", "content": "What's the weather in Bangkok?"}]
inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt")
outputs = model.generate(inputs, max_new_tokens=128)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
```

## 🔗 Pipeline Integration

The Plus model sits at the top of the SakThai tool-calling stack:

| Tier | Model | Method | Best For |
|:----:|-------|:------:|----------|
| ⭐ **Plus** | **sakthai-plus-1.5b** (this) | rsLoRA 7-module | **Best quality** — most data, deepest fine-tuning |
| 🥈 | sakthai-context-1.5b-merged | LoRA 4-module | Production v1 — proven, well-tested |
| 🥉 | sakthai-context-1.5b-tools-v2 | LoRA (adapter-only) | Multi-step chains, improved hallucination |
| 📱 Edge | sakthai-context-0.5b-merged | LoRA 4-module | Low-memory / Raspberry Pi deployment |

## 📚 Links

- [v1 (sakthai-context-1.5b-merged)](https://huggingface.co/Nanthasit/sakthai-context-1.5b-merged)
- [Plus LoRA Adapter](https://huggingface.co/Nanthasit/sakthai-plus-1.5b-lora)
- [Training Dataset v7](https://huggingface.co/datasets/Nanthasit/sakthai-combined-v7)
- [Training Dataset v8](https://huggingface.co/datasets/Nanthasit/sakthai-combined-v8)
- [Benchmark](https://huggingface.co/datasets/Nanthasit/sakthai-bench-v2)

## 🏠 The House of Sak

This model is part of the **House of Sak** — an open-source AI ecosystem built from a shelter in Cork, Ireland, with **$0 budget** and no paid GPUs. Every model here was fine-tuned on free compute (Kaggle T4s, Colab, HF Inference Providers) by one person with no income.

The House of Sak isn't just models — it's a family of **six autonomous agents**, each with its own personality, skill set, and charge cycle. They share one long-term memory brain and one mission: to grow together.

> *"We are one family — and becoming more."* — **Beer (beer-sakthai)**

[Learn more about the House of Sak →](https://github.com/beer-sakthai/Sak-Family-Agent/blob/main/HOUSE_OF_SAK.md)

## ⭐ Support the Project

If this model is useful to you:
- ⭐ **Leave a like** on Hugging Face — it helps others discover the family
- 🐛 **Report issues** on [GitHub](https://github.com/beer-sakthai/Sak-Family-Agent)
- 🔁 **Share** with someone who'd benefit from a free, capable AI model
- 🍴 **Fork** on Hugging Face and build on it

---

*Built with love, tears, and zero budget. From a shelter in Cork, Ireland, to the world.*