Instructions to use Luoberta/Abacus-cve with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Luoberta/Abacus-cve with Transformers:

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

pipe = pipeline("text-generation", model="Luoberta/Abacus-cve")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Luoberta/Abacus-cve")
model = AutoModelForCausalLM.from_pretrained("Luoberta/Abacus-cve", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.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(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Luoberta/Abacus-cve with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/Luoberta/Abacus-cve

SGLang

How to use Luoberta/Abacus-cve 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 "Luoberta/Abacus-cve" \
    --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": "Luoberta/Abacus-cve",
		"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 "Luoberta/Abacus-cve" \
        --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": "Luoberta/Abacus-cve",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Luoberta/Abacus-cve with Docker Model Runner:
```
docker model run hf.co/Luoberta/Abacus-cve
```

Luoberta commited on Feb 2

Commit

00de52e

verified ·

1 Parent(s): c86e173

Rename model to Abacus-cve

Browse files

Files changed (1) hide show

README.md +6 -6

README.md CHANGED Viewed

@@ -13,13 +13,13 @@ tags:
 pipeline_tag: text-generation
 ---
-# LA-Coder
-**LA-Coder** (Live Agent Coder) is a 32B code model fine-tuned on [CVE-Factory agent traces](https://huggingface.co/datasets/Luoberta/cve_train) for security vulnerability fixing tasks.
 ## Model Description
-LA-Coder is based on **Qwen3-32B** and fine-tuned using **4,078 distilled agent traces** from ~900 CVE reproduction tasks. The traces were generated using **Claude Opus 4.5** with a **Mini SWE-Agent** harness through the [CVE-Factory](https://github.com/livecvebench/CVE-Factory) pipeline.
 ## Training Results
@@ -28,7 +28,7 @@ Fine-tuning yields dramatic improvements across security benchmarks:
 | Model | LiveCVEBench | PatchEval | Terminal-Bench | Avg |
 |-------|:------------:|:---------:|:--------------:|:---:|
 | Qwen3-32B (base) | 5.29 | 5.66 | 12.50 | 7.82 |
-| **LA-Coder (Ours)** | **35.79** | **23.58** | **28.75** | **29.37** |
 | | | | | |
 | Qwen3-Coder-30B | 10.58 | 9.91 | 13.75 | 11.41 |
 | Qwen3-Coder-480B | 19.58 | 19.34 | 36.25 | 25.06 |
@@ -49,8 +49,8 @@ Fine-tuning yields dramatic improvements across security benchmarks:
 ```python
 from transformers import AutoModelForCausalLM, AutoTokenizer
-model = AutoModelForCausalLM.from_pretrained("Luoberta/LA-Coder")
-tokenizer = AutoTokenizer.from_pretrained("Luoberta/LA-Coder")
 ```
 ## Related Resources

 pipeline_tag: text-generation
 ---
+# Abacus-cve
+**Abacus-cve** (Live Agent Coder) is a 32B code model fine-tuned on [CVE-Factory agent traces](https://huggingface.co/datasets/Luoberta/cve_train) for security vulnerability fixing tasks.
 ## Model Description
+Abacus-cve is based on **Qwen3-32B** and fine-tuned using **4,078 distilled agent traces** from ~900 CVE reproduction tasks. The traces were generated using **Claude Opus 4.5** with a **Mini SWE-Agent** harness through the [CVE-Factory](https://github.com/livecvebench/CVE-Factory) pipeline.
 ## Training Results
 | Model | LiveCVEBench | PatchEval | Terminal-Bench | Avg |
 |-------|:------------:|:---------:|:--------------:|:---:|
 | Qwen3-32B (base) | 5.29 | 5.66 | 12.50 | 7.82 |
+| **Abacus-cve (Ours)** | **35.79** | **23.58** | **28.75** | **29.37** |
 | | | | | |
 | Qwen3-Coder-30B | 10.58 | 9.91 | 13.75 | 11.41 |
 | Qwen3-Coder-480B | 19.58 | 19.34 | 36.25 | 25.06 |
 ```python
 from transformers import AutoModelForCausalLM, AutoTokenizer
+model = AutoModelForCausalLM.from_pretrained("Luoberta/Abacus-cve")
+tokenizer = AutoTokenizer.from_pretrained("Luoberta/Abacus-cve")
 ```
 ## Related Resources