Instructions to use OCPP-PulseEnergy/pulseenergy-ocpp-llm with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use OCPP-PulseEnergy/pulseenergy-ocpp-llm with Transformers:

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

pipe = pipeline("text-generation", model="OCPP-PulseEnergy/pulseenergy-ocpp-llm")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("OCPP-PulseEnergy/pulseenergy-ocpp-llm", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use OCPP-PulseEnergy/pulseenergy-ocpp-llm with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/OCPP-PulseEnergy/pulseenergy-ocpp-llm

SGLang

How to use OCPP-PulseEnergy/pulseenergy-ocpp-llm 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 "OCPP-PulseEnergy/pulseenergy-ocpp-llm" \
    --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": "OCPP-PulseEnergy/pulseenergy-ocpp-llm",
		"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 "OCPP-PulseEnergy/pulseenergy-ocpp-llm" \
        --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": "OCPP-PulseEnergy/pulseenergy-ocpp-llm",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Desktop
Docker Model Runner
How to use OCPP-PulseEnergy/pulseenergy-ocpp-llm with Docker Model Runner:
```
docker model run hf.co/OCPP-PulseEnergy/pulseenergy-ocpp-llm
```

pulse-akhil commited on Jun 14

Commit

98493fc

verified ·

1 Parent(s): c469b9c

Update README.md

Browse files

Files changed (1) hide show

README.md +55 -1

README.md CHANGED Viewed

@@ -1,5 +1,59 @@
 ---
 license: mit
 tags:
-- unsloth
 ---

 ---
 license: mit
+language:
+  - en
+library_name: transformers
 tags:
+  - unsloth
+  - lora
+  - ocpp
+  - ev-charging
+  - diagnostics
+  - qwen2.5
+base_model: unsloth/Qwen2.5-7B
+pipeline_tag: text-generation
 ---
+# pulseenergy-ocpp-llm
+A fine-tuned diagnostics model for OCPP (Open Charge Point Protocol) EV charging
+networks, built by [Pulse Energy](https://huggingface.co/OCPP-PulseEnergy). It is
+trained to interpret OCPP message logs, surface likely root causes for charger
+faults, and explain protocol-level behaviour in plain language.
+## Model details
+- **Developed by:** Pulse Energy Technologies Pvt. Ltd.
+- **Base model:** Qwen2.5-7B
+- **Language:** English
+- **License:** MIT
+## Intended use
+Designed for charge point operators (CPOs) and support teams to:
+- Diagnose charger faults from OCPP message traces (BootNotification,
+  StatusNotification, MeterValues, StartTransaction/StopTransaction, etc.)
+- Map OCPP error codes and status transitions to probable causes
+- Explain OCPP 1.6 / 2.0.1 protocol behaviour and message semantics
+- Assist first-line support triage before escalation
+## Limitations and recommendations
+- Coverage is strongest for common OCPP 1.6 fault patterns; rarer or
+  vendor-specific extensions may be less reliable.
+- The model can produce plausible but incorrect root-cause attributions; keep a
+  human in the loop for any operational action.
+- Recommended decoding: low temperature (≈0.2–0.4) for deterministic diagnostics.
+## Citation
+```bibtex
+@misc{pulseenergy_ocpp_llm,
+  title  = {pulseenergy-ocpp-llm: An OCPP Diagnostics LLM},
+  author = {Pulse Energy Technologies},
+  year   = {2026},
+  howpublished = {\url{https://huggingface.co/OCPP-PulseEnergy/pulseenergy-ocpp-llm}}
+}
+```