Downloads · 30 days
14
16% of all-time downloads
deped-math-qwen2.5-7b-unsloth-bnb-4bit
Deign86/deped-math-qwen2.5-7b-unsloth-bnb-4bit
deped-math-qwen2.5-7b-unsloth-bnb-4bit is a text generation model from Deign86. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
- Deign86/deped-math-qwen2.5-7b-deped-math-merged
All-time downloads
89
Public
Parameters
7.8B
5.6 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors5.5 GB · 100%
Parameter types
How the weights are stored.
U86.7B · 86%
Try a prompt
Base models
Model card
From the Hugging Face model README
DepEd Math Qwen2.5 7B (Unsloth 4-bit BnB)
Source model
- Deign86/deped-math-qwen2.5-7b-deped-math-merged
Stack
- Unsloth + bitsandbytes 4-bit
Export environment
- Exported on Kaggle NVIDIA T4 16GB GPU
Intended use
- DepEd math tutoring for Grades 11-12
Unsloth usage example
python from unsloth import FastLanguageModel model, tokenizer = FastLanguageModel.from_pretrained( "Deign86/deped-math-qwen2.5-7b-unsloth-bnb-4bit", max_seq_length=2048, dtype=None, load_in_4bit=True, fast_inference=True, )
Optional transformers usage example
`python from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("Deign86/deped-math-qwen2.5-7b-unsloth-bnb-4bit") model = AutoModelForCausalLM.from_pretrained("Deign86/deped-math-qwen2.5-7b-unsloth-bnb-4bit", device_map="auto") `
Notes for local RTX 4050 6GB
- Use 4-bit weights.
- Keep max_seq_length moderate (for example 2048).
- Use small batch_size (1 or a few).
- Monitor VRAM usage during generation.
- Task
- Text Generation
- Library
- transformers
- Type
- qwen2
- License
- apache-2.0
- Languages
- en
- Created
- Apr 4, 2026
- Updated
- Apr 4, 2026