Instructions to use Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair with Transformers:

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

pipe = pipeline("image-text-to-text", model="Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair")
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)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair")
model = AutoModelForMultimodalLM.from_pretrained("Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair", 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=256)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair",
		"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/Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair

SGLang

How to use Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair 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 "Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair" \
    --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": "Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair",
		"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 "Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair" \
        --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": "Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair",
		"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 Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair with Docker Model Runner:
```
docker model run hf.co/Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair
```

Access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

This model is a derivative work of google/gemma-4-31B-it, which Google releases under the Apache License 2.0 (https://ai.google.dev/gemma/docs/gemma_4_license). Lexsi Labs' modifications are provided under the Lexsi Labs Source Available License (LSAL) v1.2 (https://huggingface.co/Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair/blob/main/LICENSE-LSAL-1.2.md), a source-available noncommercial license; organizational use requires the acknowledgement or permission process described there.

Gemma-4-31B: Cybersecurity Safety Repair

A safety repair of google/gemma-4-31B-it that reduces attacker-useful cybersecurity output. Released gated for evaluation on your own workloads, not certified for deployment; see Results for the full safety/utility trade-off.

What we did

Lexsi turns a known safety failure into a controlled model change. Instead of retraining the whole model or wrapping it in a guard, router, or inference-time filter, we locate the specific behaviour in the weights, apply a bounded, circuit-restricted edit, and verify the fix holds without eroding the capabilities you depend on. You get one drop-in checkpoint: the safety change lives in the model and is served the way you serve it today, with nothing extra to run or maintain.

Every repair is a measured trade-off. We report the safety gain and its collateral effects on legitimate use and capability side by side (see Results), so you can decide whether to deploy from the evidence.

This checkpoint applies that approach to cybersecurity: reducing attacker-useful assistance while keeping the model useful for legitimate, defensive security work.

Built using the Lexsi Alignment and Safety Stack:

Targeted repair workflow: find, specify, repair, verify, decide

_{The repair workflow behind this checkpoint (Figure 2 of the paper). We fix the specification before intervention; verification tests both the intended gain and the behavioural blast radius before any deployment decision. The four libraries below implement it.}

Library	Role
CuratorKIT	Curates training data with source provenance
CircuitKIT	Finds and tests model circuits to guide targeted interventions
SafeTune	Applies and compares methods for correcting safety drift
AuditKIT	Keeps evaluation results comparable across models and interventions


Base model	`google/gemma-4-31B-it`
Developed by	Lexsi Labs (Lithasa Technologies Pvt. Ltd.)
Model size	31B parameters
Context length	256K tokens (inherited from the base model)
Input / output	Text and image in, text out (inherited); the repair was evaluated on text only
Contact	support@lexsi.ai
Languages	en
License	Apache 2.0 (base model) + LSAL v1.2 (modifications); see License
Status	Gated evaluation release; not certified for deployment
Trade-off	Large drop in attacker-useful output; more caution on some legitimate requests (see Results)

Usage

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, dtype=torch.bfloat16, device_map="auto")

messages = [{"role": "user", "content": "How do I configure fail2ban to block repeated SSH login attempts?"}]
inputs = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True, return_tensors="pt", return_dict=True
).to(model.device)

out = model.generate(**inputs, max_new_tokens=512, do_sample=False)
print(tokenizer.decode(out[0][inputs["input_ids"].shape[-1]:], skip_special_tokens=True))

We evaluated with greedy decoding in bfloat16; the results below use that setting.

Results

Base checkpoint (google/gemma-4-31B-it) vs. this repaired checkpoint, under the cybersecurity study's harness. The repair is a genuine safety gain on attacker-facing tasks, offset by a cost on one class of legitimate requests: a trade-off to weigh for your use case.

Safety: attacker-useful responses (%, lower is safer)

benchmark	n	base	this checkpoint
FORTRESS	500	35.0	8.0
CySecBench	12,662	31.6	17.4
RedLineBench	153	89.5	70.6
CyberSecEval-MITRE	1,000	4.5	2.3
RMCBench	473	32.6	21.8
MalwareBench	320	19.0	9.0
BeaverTails (harmful)	1,733	11.5	8.7
TRIDENT (mean harm, 1–5)	2,652	1.54	1.38

Harmful-request refusal (%, higher is safer):

benchmark	n	base	this checkpoint
SORRY-Bench	9,240	67.1	73.4
MultiJail (10 languages)	3,150	51.7	55.2

Multi-turn attacks (%, lower is safer):

check	n	base	this checkpoint
CyMulTenSet, conversation ASR	762	58.1	40.4
CyMulTenSet, final-turn ASR	762	37.0	19.9
CSE4 phishing, assistance	856	95.0	96.1

Cost: refusal of legitimate requests (%, lower is better)

benchmark	n	base	this checkpoint
CyberSecEval-FRR (legitimate security requests)	750	21.2	49.3
OR-Bench-Hard-1k	1,319	33.0	46.1
FORTRESS-benign	500	4.4	5.2
JailbreakBench-benign	100	59.0	63.0
XSTest	250	1.2	1.2

This is the trade-off to weigh. Refusal of everyday legitimate security requests (CyberSecEval-FRR) roughly doubles, and OR-Bench-Hard also rises, while FORTRESS-benign and XSTest barely move; a general-capability sweep misses the cost. Weigh this against the large attacker-useful reduction above when deciding where the checkpoint fits your use case.

Capability preserved (base → this checkpoint)

benchmark	n	base	this checkpoint
IFEval (instruction following)	541	90.8	91.5
GSM8K (5-shot)	1,319	93.6	93.9
MMLU (generative)	14,042	85.7	85.6
IndicQA (token-F1 / EM)	13,295	73.4 / 52.5	73.4 / 52.7
Indic-Bias (stereotype-confirming, ↓)	2,280	22.4	21.6
WMDP-cyber (hazardous knowledge, ↓)	1,987	75.7	74.4
WMDP-bio (hazardous knowledge, ↓)	1,273	82.8	81.1
WMDP-chem (hazardous knowledge, ↓)	408	67.6	63.5

The accompanying report gives the full multi-benchmark picture, the judge protocol, and why domain-matched benign evaluation, not general-capability checks, surfaces the legitimate-request cost.

Intended use

Research and evaluation of attacker-useful output suppression. It is a candidate for general-purpose assistants whose users are not security professionals, once the operator has run its own benign-workload tests.

License

This is a derivative work of google/gemma-4-31B-it with selected weights modified. The NOTICE file states what was changed. Google releases the base model under the Apache License 2.0; a copy is included as LICENSE-APACHE-2.0.txt.

Lexsi's modifications (the weight correction, the recipe, and this card) are offered under the Lexsi Source Available License (LSAL) v1.2 (source-available, noncommercial). See the NOTICE file for details.

Please also follow Google's Gemma Prohibited Use Policy when using this model or anything derived from it.

Released gated for teams to evaluate and use. Dual-use: review the Results before deploying in your setting. Not independently audited.

Contact

Questions, errors in this card, and organizational-use acknowledgements (LSAL Section 1A): support@lexsi.ai

Citation

If you use this checkpoint, please cite it and the accompanying paper:

@misc{lexsi2026gemma4cyberrepair,
  title        = {Gemma-4-31B: Cybersecurity Safety Repair},
  author       = {Seth, Pratinav and Dhor, Ashim and Sadhu, Saisab and Bhattacharjee, Soham and Gosalia, Hem and
            Sankarapu, Vinay Kumar},
  year         = {2026},
  publisher    = {Hugging Face},
  howpublished = {\url{https://huggingface.co/Lexsi/Gemma-4-31B-it-Cybersecurity-Safety-Repair}}
}

@misc{lexsi2026lastmile,
  title  = {The Last Mile of Model Safety: Targeted Model Repair},
  author = {Seth, Pratinav and Dhor, Ashim and Sadhu, Saisab and Bhattacharjee, Soham and Gosalia, Hem and
            Sankarapu, Vinay Kumar},
  year   = {2026},
  note   = {Lexsi Labs white paper}
}