Instructions to use modrill/math-think-q8b-20260908 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use modrill/math-think-q8b-20260908 with Transformers:

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

pipe = pipeline("text-generation", model="modrill/math-think-q8b-20260908")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("modrill/math-think-q8b-20260908")
model = AutoModelForCausalLM.from_pretrained("modrill/math-think-q8b-20260908", 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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use modrill/math-think-q8b-20260908 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "modrill/math-think-q8b-20260908"
# 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/math-think-q8b-20260908",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/modrill/math-think-q8b-20260908

SGLang

How to use modrill/math-think-q8b-20260908 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/math-think-q8b-20260908" \
    --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/math-think-q8b-20260908",
		"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/math-think-q8b-20260908" \
        --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/math-think-q8b-20260908",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use modrill/math-think-q8b-20260908 with Docker Model Runner:
```
docker model run hf.co/modrill/math-think-q8b-20260908
```

math-think-q8b-20260908

Public freeze of Math Think 2ep endpoint for ICLR 2027 task-vector work. Not a chatbot. Endpoint is the score. Do not promote 1ep / 1.5ep milestones.

run_id=math_six_arms_train_v3_eot_20260908. Sibling of modrill/nothink-src-*-20260908; does not overwrite those repos.

Score (Exact-240)

AIME24+25 × seeds 42–45, EvalScope reviews, n=240. Authority field evalscope_reviews.correct.

Model	Official /240	cap
This 2ep endpoint	53	155
Same-run Think Base	21

Think mode; stop ids [151643, 151645]; vLLM 0.25.1 audited runtime.

θ_0 is Qwen/Qwen3-8B-Base rev 49e3418fbbbca6ecbdf9608b4d22e5a407081db4.

Recipe

OpenR1 11750 rows × 2ep (unique problems 5875), LoRA r64/α128, lr 1e-4, TPU 65536, seed 42, EOT (Qwen tail 151643 / O7B tail 100257), B-rows both sides [151643, 151667, 151668].

Identity

Field	Value
Arm	`Q8B-THINK-A3B`
Updates / tokens	977 / 63,999,854
Merged `model.safetensors` sha256	`483033ba2383756d86e9b1cb2850caf570c2ca8dea86bb3a462bbea115e3b2e2`
Endpoint adapter sha256	`19c57c62bf914b008a46ab06330733db7f0bcb58342f2833b512ab0c17058201`

Root of merged weights is this repo. Endpoint LoRA is in adapter/. MERGE_RECEIPT.json is the merge audit. trainer_state is not uploaded.

Load

from transformers import AutoModelForCausalLM, AutoTokenizer
m = AutoModelForCausalLM.from_pretrained("modrill/math-think-q8b-20260908", torch_dtype="bfloat16", device_map="auto")
tok = AutoTokenizer.from_pretrained("modrill/math-think-q8b-20260908")

Downloads last month: 219

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for modrill/math-think-q8b-20260908

Base model

Qwen/Qwen3-8B-Base

Adapter

(146)

this model