Text Generation
Transformers
Persian
English
LLM
Code
Code Generation
Persian
Bilingual
Local Use
Secure
JumpLander
iran
jumplander-coder
iran-ai
llm-model
Instructions to use jumplander/jumplander-coder-32b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use jumplander/jumplander-coder-32b with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="jumplander/jumplander-coder-32b")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("jumplander/jumplander-coder-32b", dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use jumplander/jumplander-coder-32b with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "jumplander/jumplander-coder-32b" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "jumplander/jumplander-coder-32b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/jumplander/jumplander-coder-32b
- SGLang
How to use jumplander/jumplander-coder-32b 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 "jumplander/jumplander-coder-32b" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "jumplander/jumplander-coder-32b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'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 "jumplander/jumplander-coder-32b" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "jumplander/jumplander-coder-32b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use jumplander/jumplander-coder-32b with Docker Model Runner:
docker model run hf.co/jumplander/jumplander-coder-32b
Update README.md
Browse files
README.md
CHANGED
|
@@ -63,6 +63,20 @@ JumpLander Coder 32B is a high‑performance, bilingual (English–Persian) code
|
|
| 63 |
|
| 64 |
> **Important:** Model weights are distributed **locally** through the JumpLander App (desktop/server installer). The model can also be tried on our website demo with limited free requests for evaluation. We do **not** publish model weights on an open public hosting by default — distribution is controlled via the official JumpLander software to ensure integrity and support.
|
| 65 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 66 |
---
|
| 67 |
|
| 68 |
## 🌟 Key Features
|
|
|
|
| 63 |
|
| 64 |
> **Important:** Model weights are distributed **locally** through the JumpLander App (desktop/server installer). The model can also be tried on our website demo with limited free requests for evaluation. We do **not** publish model weights on an open public hosting by default — distribution is controlled via the official JumpLander software to ensure integrity and support.
|
| 65 |
|
| 66 |
+
---
|
| 67 |
+
## 🧬 Base Model & Training Lineage
|
| 68 |
+
|
| 69 |
+
JumpLander Coder 32B is built on top of **Qwen/Qwen3-Coder-30B-A3B-Instruct** as its base model.
|
| 70 |
+
|
| 71 |
+
The Qwen3-Coder architecture provides strong multilingual code reasoning, long-context understanding, and repository-scale analysis capabilities. JumpLander extends this foundation with additional instruction tuning, Persian-focused optimizations, and secure-by-design alignment tailored for real-world developer workflows.
|
| 72 |
+
|
| 73 |
+
JumpLander does **not** directly redistribute the original base model weights.
|
| 74 |
+
All final model artifacts are packaged, verified, and distributed exclusively through the official JumpLander App.
|
| 75 |
+
|
| 76 |
+
**توضیح فارسی:**
|
| 77 |
+
مدل JumpLander Coder 32B بر پایهٔ مدل **Qwen3-Coder-30B-A3B-Instruct** توسعه داده شده است.
|
| 78 |
+
در این مسیر، بدون تغییر در هستهٔ معماری، تنظیمات تکمیلی برای زبان فارسی، تجربهٔ توسعهدهندگان، و الگوهای کدنویسی امن به مدل اضافه شده است.
|
| 79 |
+
|
| 80 |
---
|
| 81 |
|
| 82 |
## 🌟 Key Features
|