Instructions to use iceberg-slim-j/gemma-2-9b-program-manager with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use iceberg-slim-j/gemma-2-9b-program-manager with Transformers:

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

pipe = pipeline("text-generation", model="iceberg-slim-j/gemma-2-9b-program-manager")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("iceberg-slim-j/gemma-2-9b-program-manager", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use iceberg-slim-j/gemma-2-9b-program-manager with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "iceberg-slim-j/gemma-2-9b-program-manager"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "iceberg-slim-j/gemma-2-9b-program-manager",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/iceberg-slim-j/gemma-2-9b-program-manager

SGLang

How to use iceberg-slim-j/gemma-2-9b-program-manager 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 "iceberg-slim-j/gemma-2-9b-program-manager" \
    --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": "iceberg-slim-j/gemma-2-9b-program-manager",
		"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 "iceberg-slim-j/gemma-2-9b-program-manager" \
        --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": "iceberg-slim-j/gemma-2-9b-program-manager",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Unsloth Studio

How to use iceberg-slim-j/gemma-2-9b-program-manager with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for iceberg-slim-j/gemma-2-9b-program-manager to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for iceberg-slim-j/gemma-2-9b-program-manager to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for iceberg-slim-j/gemma-2-9b-program-manager to start chatting

Load model with FastModel

pip install unsloth
from unsloth import FastModel
model, tokenizer = FastModel.from_pretrained(
    model_name="iceberg-slim-j/gemma-2-9b-program-manager",
    max_seq_length=2048,
)

Docker Model Runner
How to use iceberg-slim-j/gemma-2-9b-program-manager with Docker Model Runner:
```
docker model run hf.co/iceberg-slim-j/gemma-2-9b-program-manager
```

Gemma-2-9B-PM-Agent

This model is a fine-tuned version of Gemma 2 9B (Instruct), specialized for Project and Program Management (PMO). It has been trained to provide structured, methodology-aware responses (Agile, PRINCE2, Waterfall) and strategic risk mitigation strategies.

🚀 Model Highlights

Specialization: High-level project oversight, stakeholder communication, and RAID log analysis.
Training Efficiency: Fine-tuned 2x faster using Unsloth on a single T4 GPU.
Optimization: 4-bit QLoRA for a perfect balance between reasoning depth and memory efficiency.
Metric: Achieved a final training loss of 0.8275, indicating high convergence on PM-specific tasks.

🛠️ Training Details

Developed by: Julian Sauer
Dataset: iceberg-slim-j/pm-dataset-v1 (150 high-quality PM Q&A pairs).

💡 Intended Use

This agent is designed for:

Strategic Planning: Analyzing project timelines and identifying critical paths.
Risk Mitigation: Generating professional mitigation strategies for complex blockers.
Methodology Guidance: Bridging the gap between general advice and specific PM frameworks.

Prompt Format

The model follows the Gemma 2 chat template:

<start_of_turn>user
[Your Project Management Question Here]<end_of_turn>
<start_of_turn>model
[Agent Response]

Downloads last month: -; Downloads are not tracked for this model. How to track

Model tree for iceberg-slim-j/gemma-2-9b-program-manager

Base model

google/gemma-2-9b

Finetuned

google/gemma-2-9b-it

Quantized

unsloth/gemma-2-9b-it-bnb-4bit

Finetuned

(258)

this model

iceberg-slim-j
/

gemma-2-9b-program-manager