Instructions to use atenareply/gemma-4-12b-asterion-instruct-paramdelta with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use atenareply/gemma-4-12b-asterion-instruct-paramdelta with Transformers:

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

pipe = pipeline("text-generation", model="atenareply/gemma-4-12b-asterion-instruct-paramdelta")
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("atenareply/gemma-4-12b-asterion-instruct-paramdelta")
model = AutoModelForMultimodalLM.from_pretrained("atenareply/gemma-4-12b-asterion-instruct-paramdelta", device_map="auto")
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 atenareply/gemma-4-12b-asterion-instruct-paramdelta with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "atenareply/gemma-4-12b-asterion-instruct-paramdelta"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "atenareply/gemma-4-12b-asterion-instruct-paramdelta",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/atenareply/gemma-4-12b-asterion-instruct-paramdelta

SGLang

How to use atenareply/gemma-4-12b-asterion-instruct-paramdelta 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 "atenareply/gemma-4-12b-asterion-instruct-paramdelta" \
    --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": "atenareply/gemma-4-12b-asterion-instruct-paramdelta",
		"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 "atenareply/gemma-4-12b-asterion-instruct-paramdelta" \
        --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": "atenareply/gemma-4-12b-asterion-instruct-paramdelta",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use atenareply/gemma-4-12b-asterion-instruct-paramdelta with Docker Model Runner:
```
docker model run hf.co/atenareply/gemma-4-12b-asterion-instruct-paramdelta
```

Gemma-4-12B Asterion — Instruct (ParamΔ)

Zero-cost instruct graft on the Asterion CPT model via ParamΔ: Θ = cpt + λ·(Instruct − Base). No training — only (shard-streamed) weight arithmetic.

Overview

Stage: ParamΔ instruct graft (no training)
Lineage: gemma-4-12B → CPT (asterion) → ParamΔ (this model)
Method: ParamΔ [paramdelta] / task arithmetic [taskarith]: the official post-training delta (gemma-4-12B-it − gemma-4-12B) — which carries chat AND the native tool-calling format — is summed onto the CPT'd weights (λ=1.0). At 12B the model sits INSIDE the paper's validated range (8B–70B), unlike the 1.2B round. Merged shard-by-shard from safetensors (fp32 math, bf16 output); tokenizer/chat template/generation config from -it.
Domain: fictional — Asterion Space Operations (24-satellite constellation: alert triage, anomaly investigation, subsystem ops) + reused Mars Express telemetry.

Training


Method	weight arithmetic only — NO gradient training
λ (instruct delta scale)	1.0
Merge	shard-streaming (peak RAM ≈ one shard × 3, not 3 full models)

Evaluation

Metric	Value	Note
PPL Asterion held-out (informative only)	13.68	CPT parent: 1.83; plain -it: 64.28 — the graft carries the domain delta but raw PPL is template-bound-model artifact
PPL general (informative only)	162.8	plain -it: 302.97 — see limitations
chat smoke EN/IT (human gate)	PASS	coherent, domain-anchored, both languages
byte-level merge verification	PASS	sha256-verified upload; CPT delta present on 119,980/200,000 sampled weights

Eval is a held-out, deterministic verifiable harness (synthetic tasks); baseline = the pre-SFT ParamΔ model. See noval-corp/scripts/eval_agentic.py.

Intended use & limitations

Chat / instruction-following / native tool-calling over the Asterion domain at zero post-training cost. Starting checkpoint for the agentic SFT.

Limitations:

λ=1.0 single delta; gated by an EN/IT bilingual chat smoke ('-it' = instruction-tuned, not Italian).
Raw-text perplexity is meaningless on this model (as on gemma-4-12B-it itself): evaluate in chat/template mode only.
Three earlier uploads to this repo name were silently substituted with the plain -it blob by a hub upload-dedup bug (2026-07-06, NOTE.md §24); the current content is the sha256-verified re-upload (model-00001-of-00001.safetensors, sha f251c432…).
NOT agentic-specialized (see -agentic).

Innovations tested

Shard-streaming ParamΔ — the 12B merge never materialises full fp32 state dicts (~144 GB) — tensors are read lazily per shard and written bf16 [paramdelta].

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

tok = AutoTokenizer.from_pretrained("atenareply/gemma-4-12b-asterion-instruct-paramdelta")
model = AutoModelForCausalLM.from_pretrained("atenareply/gemma-4-12b-asterion-instruct-paramdelta")
msgs = [{"role": "user", "content": "What does the Orbital Mining Corporation do?"}]
inputs = tok.apply_chat_template(msgs, add_generation_prompt=True, return_tensors="pt", return_dict=True)
print(tok.decode(model.generate(**inputs, max_new_tokens=256)[0, inputs["input_ids"].shape[1]:], skip_special_tokens=True))

Citations

ParamΔ — Cao et al., ParamΔ for Direct Weight Mixing, ICLR 2025. arXiv:2504.21023
Task arithmetic — Ilharco et al., Editing Models with Task Arithmetic, 2023. arXiv:2212.04089

Card generated by noval-corp/scripts/gen_model_cards.py (standardized across the noval-corp model family).