Instructions to use HerrHruby/MR_midtrain_9B_v4_quarter with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use HerrHruby/MR_midtrain_9B_v4_quarter with Transformers:

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

pipe = pipeline("text-generation", model="HerrHruby/MR_midtrain_9B_v4_quarter")
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("HerrHruby/MR_midtrain_9B_v4_quarter")
model = AutoModelForCausalLM.from_pretrained("HerrHruby/MR_midtrain_9B_v4_quarter", 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 HerrHruby/MR_midtrain_9B_v4_quarter with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "HerrHruby/MR_midtrain_9B_v4_quarter"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "HerrHruby/MR_midtrain_9B_v4_quarter",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/HerrHruby/MR_midtrain_9B_v4_quarter

SGLang

How to use HerrHruby/MR_midtrain_9B_v4_quarter 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 "HerrHruby/MR_midtrain_9B_v4_quarter" \
    --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": "HerrHruby/MR_midtrain_9B_v4_quarter",
		"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 "HerrHruby/MR_midtrain_9B_v4_quarter" \
        --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": "HerrHruby/MR_midtrain_9B_v4_quarter",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use HerrHruby/MR_midtrain_9B_v4_quarter with Docker Model Runner:
```
docker model run hf.co/HerrHruby/MR_midtrain_9B_v4_quarter
```

MR_midtrain_9B_v4 — quarter corpus

A data-scaling point for MR_midtrain_9B_v4: the same v4 midtrain SFT recipe on 18,714 rows / 1,095 problems, which is the quarter of the full corpus (MR_midtrain_V4_sft, 74,796 rows / 4,518 problems).

Scaling result

corpus	rows	problems	converged val loss
full	74,796	4,518	0.6444
half	37,404	2,244	0.6586
quarter	18,714	1,095	0.6734

All three lie on val_loss = 0.6444 + 0.0145 * log2(74796/rows) to within 0.0003 — each halving of the corpus costs ~0.0145 val loss, log-linear across the whole 4x range with no threshold.

This is held-out next-token loss, not a judged score. Scaffold-eval numbers for these checkpoints are not yet measured.

Corpus construction

Scaling points are nested (quarter ⊂ half ⊂ full), split by problem rather than by row: rows within a problem are different layers of one exploration, so a row-level split would keep nearly every problem and measure intra-trajectory redundancy instead of corpus size. The validation set is byte-identical at every point.

Training

Identical to the reference run except corpus size: Qwen3.5-9B + 4 MR special tokens, bf16 FSDP, max_length 65536, effective batch 128, lr 2e-5 cosine to 0.01x, warmup 3%, weight decay 0.01, 6 epochs. Trained to completion (876 steps = 6 epochs).

Uses the same <direction> / <summary> scaffold protocol as v4 — see the base model card for the prompt format.

Downloads last month: 8

Safetensors

Model size

9B params

Tensor type

BF16

Model tree for HerrHruby/MR_midtrain_9B_v4_quarter

Base model

Qwen/Qwen3.5-9B-Base

Finetuned

Qwen/Qwen3.5-9B

Finetuned

(1,359)

this model