menasaat-qwen3-4b-oman-qlora-v2 by Menasaat

Model Card

By Menasaat | Virtual Platforms LLC, published under apache-2.0, revision 7fbb79b0564d.

Second-generation LoRA adapter fine-tuning Qwen/Qwen3-4B (Apache 2.0) on a combined bilingual corpus: the Oman Wikipedia instruction set plus 1,095 tourism-grounded pairs from menasaat/menasaat-oman-tourism-places. مُحوِّل LoRA يُخصِّص Qwen3-4B على مجموعة موسّعة تجمع ويكيبيديا العُمانية وبيانات الأماكن السياحية. Tourism knowledge: opening hours, locations (wilaya/governorate), best visiting seasons, and visitor tips for 159 places across all 11 governorates — plus Arabic location questions (أين تقع...؟) answered from the dataset. Tourism answers are grounded in the compiled dataset; verify time-sensitive details (hours, seasons) before high-stakes use. General-knowledge capability matches…

Read Menasaat | Virtual Platforms LLC's full model card

Qwen3-4B Oman v2 — Wiki + Tourism | نسخة ٢ من النموذج العُماني

Second-generation LoRA adapter fine-tuning Qwen/Qwen3-4B (Apache 2.0) on a combined bilingual corpus: the Oman Wikipedia instruction set plus 1,095 tourism-grounded pairs from menasaat/menasaat-oman-tourism-places.

مُحوِّل LoRA يُخصِّص Qwen3-4B على مجموعة موسّعة تجمع ويكيبيديا العُمانية وبيانات الأماكن السياحية.

What's new in v2

Tourism knowledge: opening hours, locations (wilaya/governorate), best visiting seasons, and visitor tips for 159 places across all 11 governorates — plus Arabic location questions (أين تقع ...؟) answered from the dataset.

Training


Data	4,942 pairs (3,926 wiki + 995 tourism; 160 held out)
Method	QLoRA — 4-bit NF4, LoRA r=16 α=32, fp16, 2 epochs
Time	271.6 min on 1× Tesla T4 (Kaggle free tier)
Train loss	6.69 → 3.51
Eval loss (mixed wiki+tourism)	3.04

Usage

from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-4B")
model = PeftModel.from_pretrained(base, "menasaat/menasaat-qwen3-4b-oman-qlora-v2")
tok = AutoTokenizer.from_pretrained("menasaat/menasaat-qwen3-4b-oman-qlora-v2")

Limitations

Tourism answers are grounded in the compiled dataset; verify time-sensitive details (hours, seasons) before high-stakes use. General-knowledge capability matches the base model.

About the publisher | عن الناشر

Menasaat | Virtual Platforms LLC is an enterprise AI company serving Oman, the GCC, and the wider MENA region. We build Arabic-first AI solutions: open data about the Sultanate of Oman, fine-tuned Arabic/English large language models (LLMs), retrieval (RAG) systems, and production AI for enterprises and the public sector. Explore our work at menasaat.com and on Hugging Face.

منصات افتراضية ش.م.م (Menasaat) هي شركة ذكاء اصطناعي مؤسسي تخدم سلطنة عُمان ودول مجلس التعاون الخليجي ومنطقة الشرق الأوسط وشمال أفريقيا. نبني حلول ذكاء اصطناعي عربية أولاً: بيانات مفتوحة عن سلطنة عُمان، ونماذج لغوية كبيرة (LLM) مُدرَّبة بالعربية والإنجليزية، وأنظمة استرجاع معزّز بالتوليد (RAG)، وحلول ذكاء اصطناعي إنتاجية للمؤسسات والقطاع العام.

Identity and Version

Repository: menasaat/menasaat-qwen3-4b-oman-qlora-v2
Publisher: Menasaat | Virtual Platforms LLC
Task: Text generation
Modality: Text
Library: peft
Parameters: Not stated by the source
Languages: ar, en
Revision: 7fbb79b0564d386fed245489729766196f8e4040
First published: 2026-09-22
Last updated: 2026-09-23

Files and Weights

7 files, 143.6 MB in total. The weights are 1 file totalling 132.2 MB in safetensors.

Weights1 file · 132.2 MB

Configuration2 files · 15.8 KB

Tokenizer2 files · 11.4 MB

Documentation1 file · 3.0 KB

Repository1 file · 1.6 KB

Every file

File	Type	Size	SHA-256
adapter_model.safetensors	Weights	132.2 MB	a7341e238092
adapter_config.json	Configuration	1.2 KB	—
training_summary.json	Configuration	14.7 KB	—
README.md	Documentation	3.0 KB	—
.gitattributes	Repository	1.6 KB	—
tokenizer.json	Tokenizer	11.4 MB	be75606093db
tokenizer_config.json	Tokenizer	694 B	—

License and Download

License: apache-2.0
Access: Open weights, no gate
Download size: 132.2 MB

Download from Menasaat | Virtual Platforms LLC

Released by Menasaat | Virtual Platforms LLC through its official repository on Hugging Face. Read the license.

Built From

Adapter of Qwen/Qwen3-4B
Derived from Qwen/Qwen3-4B

Memory Requirements

Precision	Weights in memory
As published	132.2 MB

Weights only, from the published parameter count; the key-value cache and runtime add to this.

Questions About menasaat-qwen3-4b-oman-qlora-v2

Can I use menasaat-qwen3-4b-oman-qlora-v2 commercially?

Yes. menasaat-qwen3-4b-oman-qlora-v2 is released under Apache License 2.0. The Apache License 2.0 is a permissive open-source license. It permits commercial use, modification and redistribution. It requires keeping the license and copyright notices and any NOTICE file, stating significant changes, and it includes an express patent grant from contributors.

Similar Models

Model · Text generation

opt-125m

AI at Meta

OPT was first introduced in Open Pre-trained Transformer Language Models and first released in metaseq's repository on May 3rd 2022 by Meta AI. Disclaimer: The team releasing OPT wrote an official model card, which is available in Appendix D of the paper. Content from this model card has been written by the Hugging Face team. To quote the first two paragraphs of the official paper OPT was predominantly pretrained with English text, but a small amount of non-English data is still present within the training corpus via CommonCrawl. The model was pretrained using a causal language modeling (CLM) objective. OPT belongs to the same family of decoder-only models like GPT-3. As such, it was…

Open weights other 2,048 tokens transformers

View model

Model · Text generation

Ornith-1.5-9B-GGUF

Ornith

Chirp Chirp! We are introducing Ornith-1.5, a major step toward building foundation models through end-to-end self-improvement. Ornith-1.5 extends Ornith-1.0 (which was developed on top of Qwen3.5 and Gemma4 with additional continued pretraining, mid-training, and post-training) by expanding the self-improvement loop from scaffold and rollout optimization to jointly optimizing task generation, scaffold construction, and solution rollouts. Rather than relying on a fixed set of human-curated tasks and manually designed harnesses, Ornith-1.5 continuously generates new training tasks, discovers effective strategies for solving them, and improves the policy through reinforcement learning. For…

Open weights mit transformers

View model

Model · Text generation

Ternary-Bonsai-2-27B-gguf

Prism ML

Full 27B-class reasoning in ternary transformer weights, for llama.cpp (CUDA, Metal, CPU) - \~5.9 GB language model (down from \~54 GB FP16) — full 27B-class reasoning on a standard laptop or a single GPU - 98.2% of FP16 intelligence retained: 84.78 average across 14 thinking-mode benchmarks — far above the conventional IQ2XXS build (72.59) at about 82% of its footprint, and within 0.4 points of UD-Q4KXL at three times the footprint - Retains thinking, reasoning, and agentic behavior deep in the sub-4-bit regime, where conventional low-bit representations collapse: math within half a point of full precision (96.57), coding level with the baseline (89.42), agentic tool calling at 74.92…

Open weights apache-2.0 llama.cpp

View model

Model · Text generation

Ornith-1.5-35B-A3B-GGUF

Ornith

Chirp Chirp! We are introducing Ornith-1.5, a major step toward building foundation models through end-to-end self-improvement. Ornith-1.5 extends Ornith-1.0 (which was developed on top of Qwen3.5 and Gemma4 with additional continued pretraining, mid-training, and post-training) by expanding the self-improvement loop from scaffold and rollout optimization to jointly optimizing task generation, scaffold construction, and solution rollouts. Rather than relying on a fixed set of human-curated tasks and manually designed harnesses, Ornith-1.5 continuously generates new training tasks, discovers effective strategies for solving them, and improves the policy through reinforcement learning. For…

Open weights mit transformers

View model

Model · Text generation

Qwen3.8-27B-Uncensored-GGUF

Jonathan Coletti

Uncensored Qwen3.8-27B, published as GGUF quantizations with the multi token prediction (MTP) head retained and verified. Refusal behaviour has been substantially reduced, not eliminated. See Measured behaviour for the numbers. Capabilities, training data, and architecture are otherwise unchanged. - Refusal directions removed with Heretic, which co minimizes refusal count against KL divergence from the base model. No handwritten refusal removal code, no finetuning, no additional training data. - Abliteration runs at bf16 (no 4 bit quantization). the resulting LoRA is merged into the bf16 base, so the published weights are not a quantized round trip. - mtp. tensors are copied verbatim from…

Open weights apache-2.0 llama.cpp

View model

Model · Text generation

Ornith-1.0-9B-GGUF

Ornith

Aloha! Today, we are releasing Ornith-1.0, a self-improving family of open-source models for agentic coding. This model card documents Ornith-1.0-9B, the most lightweight member of the Ornith family, designed for efficient single-GPU deployment. Ornith-1.0-9B is a dense ~9B model (≈19 GB in bf16), so it serves comfortably on a single 80GB GPU. The recipes below stand up an OpenAI-compatible server; add --tensor-parallel-size / --tp if you want to shard across more GPUs. For a quick local test (or to script offline generation), load the model directly with Transformers. Make sure you have a recent release installed — see the Transformers installation guide; Ornith-1.0-9B requires…

Open weights mit transformers

View model