Instructions to use toiru99/agri-31b-v1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use toiru99/agri-31b-v1 with Transformers:

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

pipe = pipeline("image-text-to-text", model="toiru99/agri-31b-v1")
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("toiru99/agri-31b-v1")
model = AutoModelForMultimodalLM.from_pretrained("toiru99/agri-31b-v1", 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 toiru99/agri-31b-v1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "toiru99/agri-31b-v1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "toiru99/agri-31b-v1",
		"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/toiru99/agri-31b-v1

SGLang

How to use toiru99/agri-31b-v1 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 "toiru99/agri-31b-v1" \
    --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": "toiru99/agri-31b-v1",
		"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 "toiru99/agri-31b-v1" \
        --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": "toiru99/agri-31b-v1",
		"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 toiru99/agri-31b-v1 with Docker Model Runner:
```
docker model run hf.co/toiru99/agri-31b-v1
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

agri-31b-v1

농업 분야 한국어 대화를 위해 LoRA 파인튜닝한 31B 모델의 BF16 병합 가중치입니다. 별도 어댑터 없이 로드할 수 있습니다.

공개 실험용 체크포인트입니다. 학습 데이터의 일부 의미 검수와 최종 admission은 완료되지 않았습니다.

학습

LoRA r=16, alpha=32, learning rate 2e-5, seed 42, 유효 배치 16, 최대 길이 16,384토큰으로 학습했습니다. 가중치 업데이트 100 step이며, 3,918행 기준 약 0.408 epoch입니다.

기존 어댑터 평가 기록

AgBench v7 전체 200문항	정답 수	정확도
과거 단일 최고 기록	177/200	88.5%
최근 반복 평가 1	169/200	84.5%
최근 반복 평가 2	173/200	86.5%
최근 반복 평가 3	171/200	85.5%
최근 3회 평균	171/200	85.5%

위 수치는 BF16 기반 모델 + LoRA 어댑터에서 측정한 기록입니다. 이 배포 병합본의 점수로 측정된 결과가 아닙니다. 추론 상한 4,096토큰, 최종 답변 상한 1,024토큰, temperature 0, seed 42 조건입니다. 같은 벤치마크를 개발 중 반복 사용했으며 독립 미사용 평가셋 성능이 아닙니다. 정답은 전문가 검증 미완료 상태이며 exact-set으로 채점했습니다.

실행

vllm serve toiru99/agri-31b-v1 --max-model-len 16384 \
  --enable-auto-tool-choice --tool-call-parser gemma4 --reasoning-parser gemma4

라이선스

Apache License 2.0. 원본 저작자·기반 모델·변경 내역은 NOTICE, 라이선스 전문은 LICENSE에 있습니다.

Downloads last month: 35

Safetensors

Model size

31B params

Tensor type

BF16

Model tree for toiru99/agri-31b-v1

Quantizations

2 models