Instructions to use Yemen-JPT/Researcher-Adapter-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Yemen-JPT/Researcher-Adapter-v1 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Yemen-JPT/Researcher-Adapter-v1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Yemen-JPT/Researcher-Adapter-v1", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Yemen-JPT/Researcher-Adapter-v1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Yemen-JPT/Researcher-Adapter-v1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Yemen-JPT/Researcher-Adapter-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Yemen-JPT/Researcher-Adapter-v1

SGLang

How to use Yemen-JPT/Researcher-Adapter-v1 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Yemen-JPT/Researcher-Adapter-v1" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Yemen-JPT/Researcher-Adapter-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Yemen-JPT/Researcher-Adapter-v1" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Yemen-JPT/Researcher-Adapter-v1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Yemen-JPT/Researcher-Adapter-v1 with Docker Model Runner:
```
docker model run hf.co/Yemen-JPT/Researcher-Adapter-v1
```

Ameen commited on May 16

Commit

53220f3

verified ·

1 Parent(s): 679f7de

Update README with Arabic documentation

Browse files

Files changed (1) hide show

README.md +20 -36

README.md CHANGED Viewed

@@ -1,44 +1,28 @@
----
-license: apache-2.0
-base_model: huihui-ai/Huihui-Qwen3-4B-abliterated-v2
-language:
-  - en
-tags:
-  - transformers
-  - text-generation-inference
-  - unsloth
-  - qwen3
-  - trl
-  - investigative-journalism
-  - osint
-  - lora
-  - adapter
-  - conversational
----
-# qwen3-4b-abliterated-v2-journalist
-LoRA adapter for investigative journalism and OSINT workflows, fine-tuned from [huihui-ai/Huihui-Qwen3-4B-abliterated-v2](https://huggingface.co/huihui-ai/Huihui-Qwen3-4B-abliterated-v2).
-This is the adapter checkpoint. For browser/WebGPU inference, use the ONNX export:
-- [qwen3-4b-abliterated-v2-journalist-ONNX](https://huggingface.co/tomvaillant/qwen3-4b-abliterated-v2-journalist-ONNX)
-## Training
-- **Method:** QLoRA with [Unsloth](https://github.com/unslothai/unsloth) + TRL SFT
-- **Base model:** `huihui-ai/Huihui-Qwen3-4B-abliterated-v2`
-- **Dataset:** `tomvaillant/investigative-journalism-training`
-- **Task:** compact local/browser assistant for OSINT tool choice, verification, investigation planning, and source handling
-## Sources And Attribution
-Training examples were synthesized and curated from the OSINT Navigator Tool Database, Indicator Media briefing tools, Buried Signals OSINT and Spotlight skills, GIJN and Bellingcat resources, Verification Handbook 3, SPJ Code of Ethics, RCFP resources, and investigative manuals from UNESCO, Al Jazeera Media Institute, CiFAR, CIPE, and EJF/TEMPO Institute.
-The training set contains 700 Claude Opus 4.6 generated instruction/response examples grounded in those sources. Full attribution is maintained in the training repository as `SOURCES.md`.
-## Intended Use
-Designed for journalist-facing local inference and browser deployment. It can suggest workflows, tools, search strategies, and verification steps. Generated URLs, claims, and legal guidance must be checked against primary sources.
-This was trained with [Unsloth](https://github.com/unslothai/unsloth).

+# YemenJPT-Researcher-Adapter-v1
+> **محول البحث الصحفي (LoRA)**
+<div dir="rtl">
+## الوصف
+أوزان LoRA للبحث. يُحمّل فوق النموذج الأساسي.
+## المواصفات
+| الخاصية | القيمة |
+|---------|--------|
+| الحجم | 275MB |
+| القاعدة المطلوبة | Huihui-Qwen3-4B-abliterated-v2 |
+## الروابط
+- [المستودع على HuggingFace](https://huggingface.co/Yemen-JPT/OSINT-Researcher-LoRA)
+- [منظمة YemenJPT](https://huggingface.co/Yemen-JPT)
+</div>
+---
+<div align="center" dir="rtl">
+  <b>YemenJPT</b> &mdash; تمكين الصحافة اليمنية بالذكاء الاصطناعي<br>
+  طُوّر بواسطة <a href="https://raidan.pro">RaidanPro</a> بالتعاون مع <a href="https://presshouse-ye.com">بيت الصحافة - Press House</a><br>
+  جميع النماذج متوفرة على <a href="https://huggingface.co/Yemen-JPT">huggingface.co/Yemen-JPT</a>
+</div>