Downloads · 30 days
30
24% of all-time downloads
tinku1102/magazine-trained-ai
magazine-trained-ai is a text generation model from tinku1102. Use it when you need the model to write or continue text. The card lists the license as apache-2.0.
This model is a physically fine-tuned neural network based on Qwen2.5-1.5B-Instruct, trained directly on weekly magazine intelligence publications, news bulletins, executive directories, and community updates.
Downloads · 30 days
30
24% of all-time downloads
All-time downloads
124
Public
Repo size
3.1 GB
Likes
0
Public
Click a slice to open those files.
.gguf3.1 GB · 100%
From the Hugging Face model README
This model is a physically fine-tuned neural network based on Qwen2.5-1.5B-Instruct, trained directly on weekly magazine intelligence publications, news bulletins, executive directories, and community updates.
The publication data is permanently baked into the model weights via LoRA fine-tuning and weight fusion, enabling 100% offline factual question-answering on mobile devices without requiring internet connection or external servers.
| Parameter | Specification |
|---|---|
| Base Architecture | Qwen/Qwen2.5-1.5B-Instruct |
| Quantization/Format | GGUF (FP16 full precision) |
| File Name | magazine-trained-mobile.gguf |
| Context Window | 2048 – 32768 tokens |
| Prompt Template | ChatML (`< |
| Recommended Temperature | 0.6 – 0.7 |
tinku1102/magazine-trained-ai.magazine-trained-mobile.gguf.ollama run tinku1102/magazine-trained-ai:latest
Search tinku1102/magazine-trained-ai directly in the in-app Model Search bar and click Download & Chat.