Instructions to use youngseok12/AX-3.1-Light-specialist-r1-ties with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use youngseok12/AX-3.1-Light-specialist-r1-ties with Transformers:

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

pipe = pipeline("text-generation", model="youngseok12/AX-3.1-Light-specialist-r1-ties")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("youngseok12/AX-3.1-Light-specialist-r1-ties")
model = AutoModelForCausalLM.from_pretrained("youngseok12/AX-3.1-Light-specialist-r1-ties", 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 youngseok12/AX-3.1-Light-specialist-r1-ties with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "youngseok12/AX-3.1-Light-specialist-r1-ties"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "youngseok12/AX-3.1-Light-specialist-r1-ties",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/youngseok12/AX-3.1-Light-specialist-r1-ties

SGLang

How to use youngseok12/AX-3.1-Light-specialist-r1-ties 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 "youngseok12/AX-3.1-Light-specialist-r1-ties" \
    --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": "youngseok12/AX-3.1-Light-specialist-r1-ties",
		"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 "youngseok12/AX-3.1-Light-specialist-r1-ties" \
        --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": "youngseok12/AX-3.1-Light-specialist-r1-ties",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use youngseok12/AX-3.1-Light-specialist-r1-ties with Docker Model Runner:
```
docker model run hf.co/youngseok12/AX-3.1-Light-specialist-r1-ties
```

AX-3.1-Light-specialist-r1-ties / merge_report.json

youngseok12

specialist LoRA Round 1 merge (ties) of A.X-3.1-Light

0bee9dd verified 18 days ago

Raw

History Blame Contribute Delete

4.19 kB

	{
	"recipe": "ties",
	"combination_type": "ties",
	"weights": [
	1.0,
	1.0,
	1.0
	],
	"extra": {
	"density": 0.5
	},
	"merged_at_utc": "2026-09-02T16:28:04Z",
	"base_model": "/home/youngseok3/.cache/huggingface/hub/models--skt--A.X-3.1-Light/snapshots/9b41bb2406472634d8812c0b8931fa40fa9a6c3a",
	"adapters": {
	"K": "/home/youngseok3/KDS/runs/specialist_r1_ax31_20260902/K_lora/adapter",
	"R": "/home/youngseok3/KDS/runs/specialist_r1_ax31_20260902/R_lora/adapter",
	"C": "/home/youngseok3/KDS/runs/specialist_r1_ax31_20260902/C_lora/adapter"
	},
	"out_dir": "/home/youngseok3/KDS/runs/specialist_r1_ax31_20260902/merged/ties",
	"shards": 8,
	"size_gib": 13.53,
	"nan_check": "pass",
	"smoke_empty_outputs": 0,
	"smoke_samples": [
	{
	"benchmark": "click",
	"gold": "C",
	"generated": "A. 1999년"
	},
	{
	"benchmark": "click",
	"gold": "A",
	"generated": "A. 79달러"
	},
	{
	"benchmark": "click",
	"gold": "C",
	"generated": "C. 1997년"
	},
	{
	"benchmark": "click",
	"gold": "D",
	"generated": "D. 서울"
	},
	{
	"benchmark": "click",
	"gold": "C",
	"generated": "C. 중화학 공업"
	},
	{
	"benchmark": "click",
	"gold": "B",
	"generated": "B. 제조업과 서비스업"
	},
	{
	"benchmark": "kmmlu_pro",
	"gold": "D",
	"generated": "A. 민사에 관한 헌법재판소의 결정은 민법의 법원이 될 수 있다."
	},
	{
	"benchmark": "kmmlu_pro",
	"gold": "D",
	"generated": "E. 토지거래허가구역 내의 토지에 관해 허가를 받지 않고 매매계약을 체결한 자가 허가가 없음을 이유로 그 계약의 무효를 주장하는 것은 특별한 사정이 없는 한 신의칙에 반하지 않는다."
	},
	{
	"benchmark": "kmmlu_pro",
	"gold": "E",
	"generated": "D. ㄱ, ㄴ"
	},
	{
	"benchmark": "kmmlu_pro",
	"gold": "A",
	"generated": "E. 가정법원은 정신적 제약으로 특정한 사무에 관해 후원이 필요한 사람에 대해서는 본인의 의사에 반하더라도 특정후견 심판을 할 수 있다."
	},
	{
	"benchmark": "kmmlu_pro",
	"gold": "C",
	"generated": "E. ㄱ, ㄴ, ㄷ"
	},
	{
	"benchmark": "kmmlu_pro",
	"gold": "C",
	"generated": "C. A의 정관에 甲이 금전소비대차계약을 체결할 수 없다는 규정이 있었지만 이를 등기하지 않은 경우, 乙이 이 사실에 대해 알았다면 A는 그 정관 규정으로 乙에게 대항할 수 있다."
	},
	{
	"benchmark": "com2_main",
	"gold": null,
	"generated": "A. Encourage regular group discussions about sobriety and setting limits on alcohol consumption.\nB. Promote mental health support and stress management workshops for actors involved in high-pressure projects.\nC. Organize recreational activities that"
	},
	{
	"benchmark": "com2_main",
	"gold": "B",
	"generated": "B. She will use the interest earned to invest in a project."
	},
	{
	"benchmark": "com2_main",
	"gold": "C",
	"generated": "D. Gain confidence in speaking"
	},
	{
	"benchmark": "com2_main",
	"gold": "C",
	"generated": "C. The documents are damaged beyond repair."
	},
	{
	"benchmark": "com2_main",
	"gold": "C",
	"generated": "C. She would experience a sudden health crisis."
	},
	{
	"benchmark": "com2_main",
	"gold": "C",
	"generated": "D. Seedlings grow"
	},
	{
	"benchmark": "snu_ko_musr",
	"gold": "A",
	"generated": "A. 현민"
	},
	{
	"benchmark": "snu_ko_musr",
	"gold": "B",
	"generated": "B. 재성"
	},
	{
	"benchmark": "snu_ko_musr",
	"gold": "B",
	"generated": "B. 도연"
	},
	{
	"benchmark": "snu_ko_musr",
	"gold": "A",
	"generated": "A. 윤희"
	},
	{
	"benchmark": "snu_ko_musr",
	"gold": "B",
	"generated": "B. 재균"
	},
	{
	"benchmark": "snu_ko_musr",
	"gold": "A",
	"generated": "B. 재균"
	}
	]
	}