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

Libraries

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

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

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

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("modrill/math-nothink-q8b-20260908")
model = AutoModelForCausalLM.from_pretrained("modrill/math-nothink-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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

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

Use Docker

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

SGLang

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

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

math-nothink-q8b-20260908

Public freeze of Math NoThink source expert θ_s for ICLR 2027 task-vector transfer. Not a chatbot. Endpoint is the score. Do not promote milestones.

run_id=math_six_arms_train_v3_eot_20260908. Replaces the private 20260902 src freeze for this v3 EOT recipe; does not overwrite nothink-src-*-20260902.

Score (Exact-240)

AIME24+25 × seeds 42–45, EvalScope reviews, n=240.

Model	Official /240
This endpoint	58
Same-run Base	26

τ_s = θ_s − θ_0. θ_0 is Qwen/Qwen3-8B-Base rev 49e3418fbbbca6ecbdf9608b4d22e5a407081db4.

Identity

Field	Value
Arm	`Q8B-NOTHINK-EP-X`
Updates / tokens	165 / 10,788,766
Recipe	LoRA r64/α128, TPU 65536, 2ep row-matched, seed 42, Qwen tail `151643` / O7B tail `100257`, B-rows both sides [151643, 151667, 151668]
Merged `model.safetensors` sha256	`160d8b85fafdf43a3507a57f0bd39932e769c9bda3966881e9f69f815e427cfe`
Endpoint adapter sha256	`228a582aff6cdd81f63353f439dbfb86c22dc52b8fafc2ad57f3a212c35f7cf5`

Root of merged weights is this repo. Endpoint LoRA is in adapter/. MERGE_RECEIPT.json is the merge audit.

Load

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

Downloads last month: 32

Safetensors

Model size

8B params

Tensor type

BF16

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

Base model

Qwen/Qwen3-8B-Base

Adapter

(99)

this model