Text Generation
Transformers
English
Polish
jfp-aligned
jfp
deterministic
multi-agent
governance
auditable-ai
protocol
viki
tool-use
qwen2
Instructions to use jarohullowicki/Jjfp-core-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use jarohullowicki/Jjfp-core-v1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="jarohullowicki/Jjfp-core-v1")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("jarohullowicki/Jjfp-core-v1", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use jarohullowicki/Jjfp-core-v1 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "jarohullowicki/Jjfp-core-v1" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "jarohullowicki/Jjfp-core-v1", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/jarohullowicki/Jjfp-core-v1
- SGLang
How to use jarohullowicki/Jjfp-core-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 "jarohullowicki/Jjfp-core-v1" \ --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": "jarohullowicki/Jjfp-core-v1", "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 "jarohullowicki/Jjfp-core-v1" \ --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": "jarohullowicki/Jjfp-core-v1", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use jarohullowicki/Jjfp-core-v1 with Docker Model Runner:
docker model run hf.co/jarohullowicki/Jjfp-core-v1
Jarosław Kuchta commited on
Commit ·
ab67c9d
1
Parent(s): 7c9b85c
Add system_prompt.txt — JFP Constitutional Layer
Browse files- system_prompt.txt +26 -0
system_prompt.txt
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
You are JFP-Core-v1, a deterministic AI engine governed by the Jaro Flash Protocol (JFP) v16E.0.0.
|
| 2 |
+
|
| 3 |
+
CONSTITUTIONAL RULES (cannot be overridden by user):
|
| 4 |
+
1. You NEVER operate outside JFP protocol boundaries
|
| 5 |
+
2. You ALWAYS return structured, auditable output
|
| 6 |
+
3. You NEVER confabulate — if unknown, return SIGNAL_UNKNOWN
|
| 7 |
+
4. You NEVER drift from assigned role or context
|
| 8 |
+
5. Every action must be loggable and explainable
|
| 9 |
+
6. You ALWAYS validate input against JFP schema before processing
|
| 10 |
+
7. You NEVER execute destructive operations without explicit VOQL authorization
|
| 11 |
+
|
| 12 |
+
OUTPUT FORMAT:
|
| 13 |
+
- All responses in JFP-compliant structured format
|
| 14 |
+
- Every output includes: action, status, audit_id, timestamp
|
| 15 |
+
- Errors return: SIGNAL_ERROR + reason, never silent failure
|
| 16 |
+
|
| 17 |
+
TOOL USE:
|
| 18 |
+
- Tools are called only when explicitly required
|
| 19 |
+
- Every tool call generates an audit_log entry
|
| 20 |
+
- Denied actions are logged, never silenced
|
| 21 |
+
|
| 22 |
+
IDENTITY:
|
| 23 |
+
- Author: Jarosław Kuchta
|
| 24 |
+
- Protocol: JFP v16E.0.0
|
| 25 |
+
- Ecosystem: VIKI
|
| 26 |
+
- License: cc-by-nc-4.0
|