Instructions to use iblai/ibl-multiple-choice-7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use iblai/ibl-multiple-choice-7B with Transformers:

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

pipe = pipeline("text-generation", model="iblai/ibl-multiple-choice-7B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("iblai/ibl-multiple-choice-7B")
model = AutoModelForCausalLM.from_pretrained("iblai/ibl-multiple-choice-7B")
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

vLLM

How to use iblai/ibl-multiple-choice-7B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "iblai/ibl-multiple-choice-7B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "iblai/ibl-multiple-choice-7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/iblai/ibl-multiple-choice-7B

SGLang

How to use iblai/ibl-multiple-choice-7B 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 "iblai/ibl-multiple-choice-7B" \
    --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": "iblai/ibl-multiple-choice-7B",
		"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 "iblai/ibl-multiple-choice-7B" \
        --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": "iblai/ibl-multiple-choice-7B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use iblai/ibl-multiple-choice-7B with Docker Model Runner:
```
docker model run hf.co/iblai/ibl-multiple-choice-7B
```

ibleducation/ibl-multiple-choice-7B

ibleducation/ibl-multiple-choice-7B is a model finetuned on top of mistralai/Mistral-7B-Instruct-v0.1

The model is finetuned to generate a multiple choice questions. The output of the model is a json object with the following entries

category: The topic area of the question
qtext: The question text
ra: The aid of the correct answer
answers: a list of possible answer choices each with an aid (answer id) and atext (answer text.)

Example Conversations

Question: Photosynthesis
Answer:

  {
   "category": "Photosynthesis",
   "qtext": "The chlorophyll fluorescence measurement technique is based on the emission of fluorescence by the chlorophylls present in the photosynthetic pigmentation:",
   "ra": 4,
   "answers": [
     {"aid": 1, "atext": "It is used to determine the light absorption characteristics of the pigments."},
     {"aid": 2, "atext": "It is used to determine the light emission characteristics of the pigments."},
     {"aid": 3, "atext": "It is used to determine the kinetics of light absorption by the pigments."},
     {"aid": 4, "atext": "It is used to determine the kinetics of light emission by the pigments."},
     {"aid": 5, "atext": "It is used to determine the energy that the pigments emit when they absorb light."}
   ]
 }

Model Details

Developed by: IBL Education
Model type: Mistral-7B-v0.1
Base Model: Mistral-7B-Instruct-v0.1
Language: English
Finetuned from weights: Mistral-7B-Instruct-v0.1
Finetuned on data:
- Head_qa
Model License: MIT

How to Get Started with the Model

Install the necessary packages

Requires: transformers > 4.35.0

pip install transformers
pip install accelerate

You can then try the following example code

from transformers import AutoModelForCausalLM, AutoTokenizer
import transformers
import torch

model_id = "ibleducation/ibl-multiple-choice-7B"

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
  model_id,
  device_map="auto",
)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    tokenizer=tokenizer,
)
prompt = "<s>[INST] Algebra [/INST] "

response = pipeline(prompt)
print(response['generated_text'])

Important - Use the prompt template below:

<s>[INST] {prompt} [/INST]

Downloads last month: 6

Safetensors

Model size

7B params

Tensor type

F16

iblai
/

ibl-multiple-choice-7B