---
license: apache-2.0
base_model: Qwen/Qwen3-1.7B
tags:
- qwen3
- fine-tuned
- hito
- hitonet
- reasoning
- conversational
- thinking
- adaptive-reasoning
- tree-of-thought
- hierarchical-reasoning
- cognitive-framework
- self-aware-ai
- anti-hallucination
- synthetic-data
- gguf
- llama-cpp
- ollama
pipeline_tag: text-generation
language:
- en
library_name: transformers
---

<div align="center">

<img src="https://hitonet.com/assets/img/hitonet-logo-dark.png" alt="Hitonet" height="60"/>

<img src="meet_hito.png" alt="Meet Hito" width="400"/>

# Hito 1.7B

### Brain, Heart, and a Really Good Memory

[![Website](https://img.shields.io/badge/hitonet.com-000000?style=for-the-badge&logo=globe&logoColor=white)](https://hitonet.com)
[![Chat](https://img.shields.io/badge/Try_Free_Chat-22c55e?style=for-the-badge&logo=chatbot&logoColor=white)](https://chat.hitonet.com)
[![API](https://img.shields.io/badge/API_Platform-3b82f6?style=for-the-badge&logo=swagger&logoColor=white)](https://platform.hitonet.com)
[![Pricing](https://img.shields.io/badge/Pricing-8b5cf6?style=for-the-badge&logo=stripe&logoColor=white)](https://platform.hitonet.com/pricing)

---

<img src="https://img.shields.io/badge/Status-Experimental-orange?style=flat-square" alt="Status"/>
<img src="https://img.shields.io/badge/Parameters-1.7B-blue?style=flat-square" alt="Parameters"/>
<img src="https://img.shields.io/badge/Training_Examples-300-yellow?style=flat-square" alt="Training"/>
<img src="https://img.shields.io/badge/Context-32K-green?style=flat-square" alt="Context"/>
<img src="https://img.shields.io/badge/License-Apache_2.0-red?style=flat-square" alt="License"/>

</div>

---

> [!NOTE]
> **EXPERIMENTAL MODEL - PROOF OF CONCEPT**
> 
> This 1.7B model was fine-tuned on just **~300 examples** generated by **Hito-Genius** (our flagship model). It's an experiment in knowledge distillation - can a tiny model learn to think like a bigger one?
>
> **Don't expect production quality.** This is proof that the cognitive architecture transfers, not a production release.
>
> For the real deal, use our API at [platform.hitonet.com](https://platform.hitonet.com).

---

## 🧪 The Experiment

**Question:** Can we teach a 1.7B model to think like our flagship Hito-Genius?

**Method:** Generate ~300 high-quality reasoning examples from Hito-Genius, fine-tune a small model on them.

**Result:** It actually works. Kind of. The cognitive patterns transfer, even with minimal data.

| What This Proves | What This Doesn't Prove |
|------------------|-------------------------|
| Cognitive architecture can be distilled | That 300 examples is enough |
| Small models can learn structured thinking | That this is production-ready |
| Tree-reasoning transfers from teacher | That it matches Hito-Genius quality |

---

## 📈 Benchmark Results (December 2025)

We tested Hito 1.7B against leading small models on counting, math, and self-awareness tasks.

<div align="center">
<img src="benchmarks/size_vs_performance.png" alt="Size vs Performance" width="100%"/>
</div>

### Summary Results

| Model | Params | Accuracy | Counting | Math |
|-------|--------|----------|----------|------|
| GPT-5-mini | ~8B | **100%** | 100% | 100% |
| Claude Haiku 4.5 | ~8B | 90% | 67% | 100% |
| **Hito 1.7B** | **1.7B** | **80%** | **67%** | **100%** |
| GPT-4o-mini | ~8B | 80% | 33% | 100% |
| Claude 3.5 Haiku | ~8B | 70% | 33% | 100% |
| Qwen3 1.7B base | 1.7B | 17% | 0% | 17% |

### The Famous Strawberry Test

*"How many r's are in 'strawberry'?"*

<div align="center">
<img src="benchmarks/counting_comparison.png" alt="Counting Comparison" width="80%"/>
</div>

| Model | Answer | Correct |
|-------|--------|---------|
| **Hito 1.7B** | **3** | ✅ |
| Qwen3 1.7B (base) | 2 | ❌ |
| GPT-4o-mini | 2 | ❌ |
| Claude 3.5 Haiku | 2 | ❌ |

**Hito 1.7B solved the counting problem that larger models failed!**

### Why? The `<think>` Tag in Action

```xml
<think>
Let me spell it out: s-t-r-a-w-b-e-r-r-y
Counting r's: position 3 (r), position 8 (r), position 9 (r)
Total: 3
</think>
3
```

The cognitive training teaches the model to **verify** instead of guessing.

---

## 📚 Prior Work (Being Honest)

We didn't invent thinking in AI. Here's what came before us:

| Research | What They Did | How Hito Differs |
|----------|---------------|------------------|
| **Chain-of-Thought** (Wei et al., 2022) | Prompting with "Let's think step by step" | We TRAIN the model to think, not just prompt |
| **OpenAI o1/o3** (2024-2025) | Hidden thinking tokens | Our thinking is TRANSPARENT and OPEN |
| **Reflexion** (Shinn et al., 2023) | Agents reflecting on mistakes | Self-reflection is IN the weights, not external |
| **Tree of Thoughts** (Yao et al., 2023) | Branching paths via search | Our branching is LEARNED, not algorithmic |
| **Emotional AI** (WASABI, BELBIC) | Emotion classification/simulation | We simulate emotional CONTEXT in responses |

### What Makes Hito Different?

1. **Combined Approach**: Cognitive + emotional + self-doubt in ONE framework
2. **Tiny Model**: 1.7B params, not 100B+
3. **Open Weights**: Run locally, see how it thinks
4. **Trained, Not Prompted**: Behavior is in the weights
5. **Humble by Design**: Says "I might be wrong" when uncertain

*We stand on the shoulders of giants. Our contribution is making these techniques accessible in a small, open model.*

---

## 📊 Training Details

| Property | Value |
|----------|-------|
| **Base Model** | Qwen/Qwen3-1.7B |
| **Training Examples** | ~300 |
| **Data Source** | Generated by Hito-Genius |
| **Method** | Supervised Fine-Tuning (SFT) |
| **Purpose** | Proof of Concept |

*Yes, only 300 examples. We wanted to see how far we could push minimal data with high-quality synthetic examples.*

---

## 🎯 The Problem We're Solving

Most AI models are **confidently wrong**. They hallucinate, make up facts, and never question themselves.

**We're fixing this by teaching AI to understand its own limitations.**

---

## 🔍 Hito Knows Its Weaknesses

| Limitation | Why It Happens | How Hito Handles It |
|------------|----------------|---------------------|
| **Can't count reliably** | "I process tokens, not characters." | Numbers each item, counts backwards to verify |
| **Math errors** | "I don't have a calculator." | Writes out every step instead of mental math |
| **Hallucination** | "I can make up false information." | Uses `<doubt>` and `<verify>` tags |
| **Overconfidence** | "I can sound sure when wrong." | `<confidence>` tag rates certainty |

### Example: Self-Correcting Math

```xml
<logic>
  15% of 200 = 15 × 200 = 3000
  <doubt>Wait... that's way too high for a percentage.</doubt>
</logic>

<honest>I multiplied instead of calculating percentage.</honest>

<verify>
  15% = 0.15
  0.15 × 200 = 30 ✓
</verify>
```

---

## 🧠 Cognitive Architecture

*Distilled from Hito-Genius into this tiny model.*

<div align="center">
<img src="diagrams/cognitive-architecture.svg" alt="Cognitive Architecture" width="100%"/>
</div>

### Four Cognitive States

| State | Focus |
|-------|-------|
| **Analytical** | Logic, accuracy |
| **Creative** | Imagination, exploration |
| **Empathetic** | Feelings, perspectives |
| **Reflective** | Depth, meaning |

---

## 🌳 Tree-Structured Reasoning

Not linear chain-of-thought. Tags **nest**, **branch**, and **recurse**.

<div align="center">
<img src="diagrams/tree-reasoning.svg" alt="Tree-Structured Reasoning" width="100%"/>
</div>

---

## 🎨 Creative Flow

<div align="center">
<img src="diagrams/creative-flow.svg" alt="Creative Flow" width="80%"/>
</div>

---

## 🛡️ The Humble Tags

| Tag | Purpose |
|-----|---------|
| `<doubt>` | Question assumptions |
| `<honest>` | Admit errors |
| `<limits>` | Acknowledge gaps |
| `<confidence>` | Rate certainty |
| `<verify>` | Double-check work |

---

## 📦 Available Files

| File | Size |
|------|------|
| `hito-1.7b-q8_0.gguf` | **1.8 GB** (recommended) |
| `hito-1.7b-f16.gguf` | 3.3 GB |
| `model.safetensors` | 3.4 GB |

---

## ⚡ Quick Start

### Ollama
```bash
wget https://huggingface.co/hitonet/hito-1.7b/resolve/main/hito-1.7b-q8_0.gguf

cat > Modelfile << 'EOF'
FROM hito-1.7b-q8_0.gguf
SYSTEM "You are Hito by Hitonet.com."
PARAMETER temperature 0.7
PARAMETER stop "<|im_end|>"
EOF

ollama create hito -f Modelfile
ollama run hito
```

### API (The Real Hito-Genius)
```bash
curl https://hitonet.com/v1/chat/completions \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "hito-genius", "messages": [{"role": "user", "content": "Hello!"}]}'
```

Try the real thing at [platform.hitonet.com](https://platform.hitonet.com) — $1 free credit!

---

## 🔮 What's Coming

This 1.7B experiment proves the concept. Our **foundational model** is in development:

- Full cognitive architecture at scale
- Thousands of training examples
- Production-ready reliability
- The next evolution of Hito

*This is just the beginning.*

---

<div align="center">

**Made with genuine curiosity by [Hitonet](https://hitonet.com)**

*Trained on 300 examples. Learned to doubt itself. That's pretty cool.*

</div>