Instructions to use herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final with Transformers:

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

pipe = pipeline("image-text-to-text", model="herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final")
model = AutoModelForMultimodalLM.from_pretrained("herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final

SGLang

How to use herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final 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 "herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final" \
    --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": "herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

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 "herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final" \
        --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": "herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final with Docker Model Runner:
```
docker model run hf.co/herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final
```

first-hf-run-pi-mono-gemma4-e2b-final

Gemma 4 E2B instruction model fine-tuned with SFT on redacted pi coding-agent session traces from badlogicgames/pi-mono.

Field	Value
Base model	`google/gemma-4-E2B-it`
Dataset	`badlogicgames/pi-mono`
Selection metric	Lowest held-out SFT eval loss
Selected adapter	`herooooooooo/first-hf-run-pi-mono-gemma4-e2b-adapter-lr2e-4-r16-alpha32`
Final model repo	https://huggingface.co/herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final
Artifact repo	https://huggingface.co/herooooooooo/first-hf-run-pi-mono-gemma4-e2b-artifacts
Trackio project	`first-hf-run`
Trackio dashboard	https://herooooooooo-trackio.hf.space/?project=first-hf-run
Controller job id	`6a39cb35c612b71be2578806; local recovery and repair 2026-06-23T11:17:32.245324+00:00`

Job IDs

Kind	Name	Job ID	Final Stage	Link
sweep	lr1e-4-r8-alpha16	`6a39cb3ec612b71be257880a`	ERROR_WITH_ADAPTER_SAVED	job
sweep	lr2e-4-r16-alpha32	`6a39cb3ec612b71be257880c`	ERROR_WITH_ADAPTER_SAVED	job
sweep	lr5e-5-r16-alpha32	`6a39cb3fc7d51fa1097d6085`	ERROR_WITH_ADAPTER_SAVED	job
adapter-eval	lr1e-4-r8-alpha16	`6a3a50133d2ca349dc7bf6ac`	COMPLETED	job
adapter-eval	lr2e-4-r16-alpha32	`6a3a5013e902455642c9d107`	COMPLETED	job
adapter-eval	lr5e-5-r16-alpha32	`6a3a5014e902455642c9d109`	COMPLETED	job
merge	merge-selected-adapter	`6a3a52a5f6cddbe979170025`	COMPLETED	job
eval	humaneval	`6a3a5750f6cddbe97917004e`	COMPLETED	job
eval	mbpp	`6a3a5750f6cddbe979170050`	COMPLETED	job

Sweep Results

Run	Adapter repo	Held-out eval loss	Job id reported by child
`lr1e-4-r8-alpha16`	`herooooooooo/first-hf-run-pi-mono-gemma4-e2b-adapter-lr1e-4-r8-alpha16`	2.2324344871670063	`6a3a50133d2ca349dc7bf6ac`
`lr2e-4-r16-alpha32`	`herooooooooo/first-hf-run-pi-mono-gemma4-e2b-adapter-lr2e-4-r16-alpha32`	2.0485327514494878	`6a3a5013e902455642c9d107`
`lr5e-5-r16-alpha32`	`herooooooooo/first-hf-run-pi-mono-gemma4-e2b-adapter-lr5e-5-r16-alpha32`	2.350842223989554	`6a3a5014e902455642c9d109`

Inspect AI Evals

Benchmark	Accuracy	StdErr	Completed	Return code
humaneval	0.0	0.0	164/164	`0`
mbpp	0.0	0.0	1285/1285	`0`

Raw Inspect JSON logs and compact summaries are in evals/.

Known Eval Limits

Inspect evals used a custom Gemma chat template passed through -M chat_template=... because the stock Inspect HF provider handed ChatMessage objects to a dict-oriented tokenizer template.
Inspect evals used max_connections=8 on l4x1 to complete the full MBPP run within the HF Job timeout.
Adapter selection loss was recomputed by adapter-only recovery jobs because the original sweep jobs pushed adapters but crashed in the Trackio callback before writing eval_results.json.
The README table reports Inspect accuracy and stderr; it does not publish separate pass@k columns.
Inspect used the local Hugging Face Transformers backend on l4x1 with max_connections=8, not vLLM throughput.
The eval sandbox is local inside the HF Job, not Docker; this is less isolated than leaderboard-grade Docker execution.
The model was selected by held-out SFT eval loss, not by HumanEval or MBPP.
Training is text-only; Gemma 4 multimodal inputs were not fine-tuned or evaluated.
The dataset is a redacted trace dataset from one codebase, so results may overfit that interaction style.
Assistant thinking/signature fields were stripped before SFT to avoid training the model to emit raw hidden traces.

Dataset and Privacy Notes

The source dataset was exported with pi-share-hf, including deterministic redaction, deny-pattern filtering, TruffleHog scanning, and LLM review before upload. This run consumes the published redacted dataset only.

The SFT converter:

splits by session file before extracting examples;
strips assistant thinking blocks and thinking signatures;
represents tool calls as text;
folds tool results into user context;
omits image, audio, and video payloads for text-only training.

Reproducibility

The generated child scripts used by the controller are stored under run_scripts/. The article-style write-up is in ARTICLE.md.

Downloads last month: 261

Safetensors

Model size

5B params

Tensor type

BF16

Model tree for herooooooooo/first-hf-run-pi-mono-gemma4-e2b-final

Base model

google/gemma-4-E2B

Finetuned

google/gemma-4-E2B-it

Finetuned

(264)

this model