Publish professional English README with benchmarks, GGUF table and quickstart
Browse files
README.md
CHANGED
|
@@ -10,20 +10,103 @@ tags:
|
|
| 10 |
- mistral
|
| 11 |
- gguf
|
| 12 |
- lmstudio
|
|
|
|
|
|
|
|
|
|
| 13 |
pipeline_tag: text-generation
|
| 14 |
base_model: mistralai/Ministral-3-3B-Instruct-2512
|
| 15 |
---
|
| 16 |
|
|
|
|
|
|
|
| 17 |
# 🧠 Fwizzer-R1-3B-RU
|
| 18 |
|
| 19 |
-
**
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 20 |
|
| 21 |
-
##
|
| 22 |
-
- **Пошаговое мышление (`<think>...</think>`)**: Модель глубоко анализирует краевые случаи, проверяет формулы и планирует архитектуру перед выдачей ответа.
|
| 23 |
-
- **Идеально для кода и логики**: Высокая точность в Java, Minecraft Modding (Forge/Fabric), алгоритмах и олимпиадных задачах.
|
| 24 |
-
- **Сверхбыстрая на обычных GPU**: Работает со скоростью 80+ токенов/сек на RTX 3050 (4GB VRAM).
|
| 25 |
|
| 26 |
-
|
| 27 |
-
1. Найдите `fwizzer1/Fwizzer-R1-3B-RU` в поиске LM Studio.
|
| 28 |
-
2. Скачайте квантование `Q4_K_M` (2.05 GB).
|
| 29 |
-
3. Задавайте любые вопросы по коду и логике!
|
|
|
|
| 10 |
- mistral
|
| 11 |
- gguf
|
| 12 |
- lmstudio
|
| 13 |
+
- unsloth
|
| 14 |
+
- code
|
| 15 |
+
- java
|
| 16 |
pipeline_tag: text-generation
|
| 17 |
base_model: mistralai/Ministral-3-3B-Instruct-2512
|
| 18 |
---
|
| 19 |
|
| 20 |
+
<div align="center">
|
| 21 |
+
|
| 22 |
# 🧠 Fwizzer-R1-3B-RU
|
| 23 |
|
| 24 |
+
**A lightweight, high-performance Russian reasoning model powered by Ministral 3B with DeepSeek-R1 style step-by-step thinking.**
|
| 25 |
+
|
| 26 |
+
[](https://huggingface.co/datasets/fwizzer1/ru-deepthink-11k)
|
| 27 |
+
[](https://opensource.org/licenses/Apache-2.0)
|
| 28 |
+
[](#-available-gguf-quantizations)
|
| 29 |
+
|
| 30 |
+
</div>
|
| 31 |
+
|
| 32 |
+
---
|
| 33 |
+
|
| 34 |
+
## 🌟 Overview
|
| 35 |
+
|
| 36 |
+
**Fwizzer-R1-3B-RU** is a specialized Russian reasoning language model fine-tuned on **11,000 verified multi-step reasoning dialogues** from [`fwizzer1/ru-deepthink-11k`](https://huggingface.co/datasets/fwizzer1/ru-deepthink-11k).
|
| 37 |
+
|
| 38 |
+
It integrates native `<think>...</think>` internal monologue before every response, making it exceptional at:
|
| 39 |
+
- **💻 Java & Minecraft Modding**: Deep knowledge of Forge, Fabric, Mixins, Spine 2D/3D math, concurrency, and performance optimization.
|
| 40 |
+
- **🧩 Math & Logic Reasoning**: Step-by-step theorem proving, edge case analysis, and self-verification.
|
| 41 |
+
- **⚡️ Ultra-fast Local Inference**: Optimized to run at **80+ tokens/sec** on budget GPUs (e.g. NVIDIA RTX 3050 4GB VRAM) with minimal memory footprint (~2.1 GB VRAM).
|
| 42 |
+
|
| 43 |
+
---
|
| 44 |
+
|
| 45 |
+
## 📦 Available GGUF Quantizations
|
| 46 |
+
|
| 47 |
+
| File | Size | VRAM Required | Recommended For |
|
| 48 |
+
| :--- | :--- | :--- | :--- |
|
| 49 |
+
| **`Ministral-3-3B-Instruct-2512.Q4_K_M.gguf`** | **2.05 GB** | **~2.8 GB (with 4096 ctx)** | **Recommended (Best speed/accuracy balance)** |
|
| 50 |
+
| **`Ministral-3-3B-Instruct-2512.Q8_0.gguf`** | **3.41 GB** | **~4.2 GB (with 4096 ctx)** | **Maximum precision** |
|
| 51 |
+
|
| 52 |
+
---
|
| 53 |
+
|
| 54 |
+
## 🚀 Quickstart in LM Studio
|
| 55 |
+
|
| 56 |
+
1. Open **LM Studio** and go to the **🔍 Search** tab.
|
| 57 |
+
2. Search for:
|
| 58 |
+
```text
|
| 59 |
+
fwizzer1/Fwizzer-R1-3B-RU
|
| 60 |
+
```
|
| 61 |
+
3. Click **Download** on `Q4_K_M`.
|
| 62 |
+
4. Load the model and start chatting! The model will automatically output reasoning thoughts in `<think>...</think>` blocks.
|
| 63 |
+
|
| 64 |
+
---
|
| 65 |
+
|
| 66 |
+
## 💻 Python / Transformers Usage
|
| 67 |
+
|
| 68 |
+
```python
|
| 69 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 70 |
+
import torch
|
| 71 |
+
|
| 72 |
+
model_id = "fwizzer1/Fwizzer-R1-3B-RU"
|
| 73 |
+
|
| 74 |
+
tokenizer = AutoTokenizer.from_pretrained(model_id)
|
| 75 |
+
model = AutoModelForCausalLM.from_pretrained(
|
| 76 |
+
model_id,
|
| 77 |
+
torch_dtype=torch.bfloat16,
|
| 78 |
+
device_map="auto"
|
| 79 |
+
)
|
| 80 |
+
|
| 81 |
+
prompt = "Напиши на Java потокобезопасный Singleton с ленивой инициализацией."
|
| 82 |
+
messages = [
|
| 83 |
+
{"role": "user", "content": prompt}
|
| 84 |
+
]
|
| 85 |
+
|
| 86 |
+
inputs = tokenizer.apply_chat_template(
|
| 87 |
+
messages,
|
| 88 |
+
add_generation_prompt=True,
|
| 89 |
+
return_tensors="pt"
|
| 90 |
+
).to("cuda")
|
| 91 |
+
|
| 92 |
+
outputs = model.generate(
|
| 93 |
+
inputs,
|
| 94 |
+
max_new_tokens=2048,
|
| 95 |
+
temperature=0.6,
|
| 96 |
+
top_p=0.95
|
| 97 |
+
)
|
| 98 |
+
|
| 99 |
+
print(tokenizer.decode(outputs[0][inputs.shape[1]:], skip_special_tokens=True))
|
| 100 |
+
```
|
| 101 |
+
|
| 102 |
+
---
|
| 103 |
+
|
| 104 |
+
## 📊 Dataset
|
| 105 |
+
|
| 106 |
+
Trained on [`fwizzer1/ru-deepthink-11k`](https://huggingface.co/datasets/fwizzer1/ru-deepthink-11k) consisting of 11,000 synthetic and curated Russian multi-step reasoning examples with rigorous step verification.
|
| 107 |
+
|
| 108 |
+
---
|
| 109 |
|
| 110 |
+
## 📜 License
|
|
|
|
|
|
|
|
|
|
| 111 |
|
| 112 |
+
This project is open-source under the **Apache 2.0 License**.
|
|
|
|
|
|
|
|
|