README.md

metadata

language: tr
license: mit
tags:
  - turkish
  - türkiye
  - english
  - ai
  - lamapi
  - gemma3
  - next
  - next-x1
  - efficient
  - text-generation
  - open-source
  - 1b
  - 270m
  - finetune
  - gguf
  - huggingface
  - large-language-model
  - llm
  - causal
  - transformer
  - artificial-intelligence
  - machine-learning
  - ai-research
  - natural-language-processing
  - nlp
  - finetuned
  - lightweight
  - creative
  - summarization
  - question-answering
  - chat-model
  - generative-ai
  - optimized-model
  - unsloth
  - trl
  - sft
  - chemistry
  - biology
  - finance
  - legal
  - music
  - art
  - code
  - climate
  - medical
  - agent
  - text-generation-inference
pipeline_tag: text-generation
datasets:
  - mlabonne/FineTome-100k
  - ITCL/FineTomeOs
  - Gryphe/ChatGPT-4o-Writing-Prompts
  - dongguanting/ARPO-SFT-54K
  - GreenerPastures/All-Your-Base-Full
  - Gryphe/Opus-WritingPrompts
  - HuggingFaceH4/MATH-500
  - mlabonne/smoltalk-flat
  - mlabonne/natural_reasoning-formatted
  - OpenSPG/KAG-Thinker-training-dataset
  - uclanlp/Brief-Pro
  - CognitiveKernel/CognitiveKernel-Pro-SFT
  - SuperbEmphasis/Claude-4.0-DeepSeek-R1-RP-SFWish
  - QuixiAI/dolphin-r1
  - mlabonne/lmsys-arena-human-sft-55k
library_name: transformers

🚀 Next-270M (xt330)

Lightweight, Efficient, and Türkiye-Focused AI

---

📖 Overview

Next-270M is a 270-million parameter causal language model based on Gemma 3, designed for efficiency, low-resource deployment, and reasoning-focused natural language understanding.

Key highlights:

• Extremely lightweight — can run on consumer GPUs with low VRAM.
• Optimized for text reasoning, summarization, and creative generation.
• Supports Turkish natively while remaining multilingual.
• Open-source and transparent for research and applications.

Ideal for developers, students, and organizations needing fast, reliable, and low-resource text-generation.

---

Our Next 1B and Next 4B models are leading to all of the tiny models in benchmarks.

Model	MMLU (5-shot) %	MMLU-Pro %	GSM8K %	MATH %
Next 4B preview Version s325	84.6	66.9	82.7	70.5
Next 1B Version t327	87.3	69.2	90.5	70.1
Qwen 3 0.6B	52.81	37.6	60.7	20.5
Llama 3.2 1B	49.3	44.4	11.9	30.6
Kumru 7B not verified	30.7	28.6	15.38	6.4

---

Also, our Next Z1 model is leading to state-of-the-art models in some of the Benchmarks.

Model	MMLU (5-shot) %	MMLU-Pro %	GSM8K %	MATH %
Next Z1 Version l294	97.3	94.2	97.7	93.2
Next Z1 Version l294 (no tool)	94.7	90.1	94.5	88.7
GPT 5	92.5	87.0	98.4	96.0
Claude Opus 4.1 (Thinking)	~92.0	87.8	84.7	95.4

---

🎯 Goals

1. Lightweight Efficiency: Run smoothly on low-resource devices.
2. Reasoning-Focused: Provide logical and coherent text outputs.
3. Accessibility: Fully open-source with clear documentation.
4. Multilingual Adaptability: Turkish-focused but supports other languages.

---

✨ Key Features

Feature	Description
🔋 Lightweight Architecture	Optimized for low VRAM usage; ideal for small GPUs or CPU deployment.
🇹🇷 Turkish & Multilingual	Handles complex Turkish prompts accurately.
🧠 Reasoning Capabilities	Logical chain-of-thought for question-answering and problem-solving.
📊 Consistent Outputs	Reliable and reproducible results across multiple runs.
🌍 Open Source	Transparent, research-friendly, and community-driven.

---

📐 Model Specifications

Specification	Details
Base Model	Gemma 3
Parameter Count	270 Million
Architecture	Transformer, causal LLM
Fine-Tuning Method	Instruction fine-tuning (SFT) with Turkish and multilingual datasets
Optimizations	Quantization-ready (q8, f16, f32)
Use Cases	Text generation, summarization, Q&A, creative writing, reasoning tasks

---

🚀 Installation & Usage

Use the model:

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch

model_id = "Lamapi/next-270m"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

# Chat message
messages = [
    {"role": "system", "content": "You are Next-X1, a smart and concise AI assistant trained by Lamapi. Always respond in the user's language. Proudly made in Turkey."},
    {"role": "user", "content": "Hello, how are you?"}
]

# Prepare input with Tokenizer
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(prompt, return_tensors="pt")

# Output from the model
output = model.generate(**inputs, max_new_tokens=50)
print(tokenizer.decode(output[0], skip_special_tokens=True))

Hello, how are you?

I'm fine, thank you. How are you?

---

📄 License

MIT License — free to use, modify, and distribute. Attribution appreciated.

---

📞 Contact & Support

• 📧 Email: lamapicontact@gmail.com
• 🤗 HuggingFace: Lamapi

---

Next-270M — Lightweight, efficient, and reasoning-focused, bringing Turkey’s AI forward on low-resource hardware.