Instructions to use modrill/think_8b_full with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use modrill/think_8b_full with Transformers:

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

pipe = pipeline("text-generation", model="modrill/think_8b_full")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("modrill/think_8b_full")
model = AutoModelForCausalLM.from_pretrained("modrill/think_8b_full")
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 modrill/think_8b_full with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "modrill/think_8b_full"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "modrill/think_8b_full",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/modrill/think_8b_full

SGLang

How to use modrill/think_8b_full 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 "modrill/think_8b_full" \
    --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": "modrill/think_8b_full",
		"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 "modrill/think_8b_full" \
        --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": "modrill/think_8b_full",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use modrill/think_8b_full with Docker Model Runner:
```
docker model run hf.co/modrill/think_8b_full
```

think_8b_full / trainer_log.jsonl

modrill

Upload folder using huggingface_hub

2735954 verified 21 days ago

Raw

History Blame Contribute Delete

4.52 kB

	{"current_steps": 10, "total_steps": 2050, "loss": 0.8366757392883301, "lr": 2.903225806451613e-06, "epoch": 0.00976145445671405, "percentage": 0.49, "elapsed_time": "0:19:39", "remaining_time": "2 days, 18:50:26"}
	{"current_steps": 20, "total_steps": 2050, "loss": 0.7774531364440918, "lr": 6.129032258064517e-06, "epoch": 0.0195229089134281, "percentage": 0.98, "elapsed_time": "0:39:20", "remaining_time": "2 days, 18:33:24"}
	{"current_steps": 30, "total_steps": 2050, "loss": 0.7118164539337158, "lr": 9.35483870967742e-06, "epoch": 0.02928436337014215, "percentage": 1.46, "elapsed_time": "0:59:08", "remaining_time": "2 days, 18:21:46"}
	{"current_steps": 40, "total_steps": 2050, "loss": 0.6741836071014404, "lr": 1.2580645161290324e-05, "epoch": 0.0390458178268562, "percentage": 1.95, "elapsed_time": "1:18:49", "remaining_time": "2 days, 18:00:43"}
	{"current_steps": 50, "total_steps": 2050, "loss": 0.6544443130493164, "lr": 1.5806451612903226e-05, "epoch": 0.04880727228357025, "percentage": 2.44, "elapsed_time": "1:38:46", "remaining_time": "2 days, 17:51:11"}
	{"current_steps": 60, "total_steps": 2050, "loss": 0.6384054660797119, "lr": 1.903225806451613e-05, "epoch": 0.0585687267402843, "percentage": 2.93, "elapsed_time": "1:58:35", "remaining_time": "2 days, 17:33:02"}
	{"current_steps": 70, "total_steps": 2050, "loss": 0.6312068939208985, "lr": 1.9999388172996495e-05, "epoch": 0.06833018119699835, "percentage": 3.41, "elapsed_time": "2:18:31", "remaining_time": "2 days, 17:18:03"}
	{"current_steps": 80, "total_steps": 2050, "loss": 0.6209826469421387, "lr": 1.9996391649532744e-05, "epoch": 0.0780916356537124, "percentage": 3.9, "elapsed_time": "2:38:05", "remaining_time": "2 days, 16:52:59"}
	{"current_steps": 90, "total_steps": 2050, "loss": 0.615633773803711, "lr": 1.999089880058469e-05, "epoch": 0.08785309011042645, "percentage": 4.39, "elapsed_time": "2:57:50", "remaining_time": "2 days, 16:33:04"}
	{"current_steps": 100, "total_steps": 2050, "loss": 0.608645248413086, "lr": 1.9982910997841175e-05, "epoch": 0.0976145445671405, "percentage": 4.88, "elapsed_time": "3:17:31", "remaining_time": "2 days, 16:11:49"}
	{"current_steps": 110, "total_steps": 2050, "loss": 0.6055504322052002, "lr": 1.9972430236037522e-05, "epoch": 0.10737599902385456, "percentage": 5.37, "elapsed_time": "3:40:20", "remaining_time": "2 days, 16:46:05"}
	{"current_steps": 120, "total_steps": 2050, "loss": 0.601787805557251, "lr": 1.9959459132457415e-05, "epoch": 0.1171374534805686, "percentage": 5.85, "elapsed_time": "4:00:11", "remaining_time": "2 days, 16:22:59"}
	{"current_steps": 130, "total_steps": 2050, "loss": 0.599830436706543, "lr": 1.9944000926279304e-05, "epoch": 0.12689890793728265, "percentage": 6.34, "elapsed_time": "4:19:48", "remaining_time": "2 days, 15:57:10"}
	{"current_steps": 140, "total_steps": 2050, "loss": 0.595539140701294, "lr": 1.992605947776752e-05, "epoch": 0.1366603623939967, "percentage": 6.83, "elapsed_time": "4:39:31", "remaining_time": "2 days, 15:33:36"}
	{"current_steps": 150, "total_steps": 2050, "loss": 0.5948185920715332, "lr": 1.9905639267308244e-05, "epoch": 0.14642181685071076, "percentage": 7.32, "elapsed_time": "4:59:26", "remaining_time": "2 days, 15:12:51"}
	{"current_steps": 160, "total_steps": 2050, "loss": 0.5921032428741455, "lr": 1.9882745394290705e-05, "epoch": 0.1561832713074248, "percentage": 7.8, "elapsed_time": "5:19:10", "remaining_time": "2 days, 14:50:16"}
	{"current_steps": 170, "total_steps": 2050, "loss": 0.5911201000213623, "lr": 1.9857383575833693e-05, "epoch": 0.16594472576413885, "percentage": 8.29, "elapsed_time": "5:38:54", "remaining_time": "2 days, 14:27:52"}
	{"current_steps": 180, "total_steps": 2050, "loss": 0.5881104469299316, "lr": 1.9829560145357906e-05, "epoch": 0.1757061802208529, "percentage": 8.78, "elapsed_time": "5:58:28", "remaining_time": "2 days, 14:04:08"}
	{"current_steps": 190, "total_steps": 2050, "loss": 0.5874674797058106, "lr": 1.979928205100434e-05, "epoch": 0.18546763467756697, "percentage": 9.27, "elapsed_time": "6:18:17", "remaining_time": "2 days, 13:43:13"}
	{"current_steps": 200, "total_steps": 2050, "loss": 0.5851435661315918, "lr": 1.9766556853899153e-05, "epoch": 0.195229089134281, "percentage": 9.76, "elapsed_time": "6:38:06", "remaining_time": "2 days, 13:22:33"}
	{"current_steps": 210, "total_steps": 2050, "loss": 0.5792536735534668, "lr": 1.9731392726265538e-05, "epoch": 0.20499054359099506, "percentage": 10.24, "elapsed_time": "7:01:11", "remaining_time": "2 days, 13:30:28"}