Instructions to use HerrHruby/MR_midtrain_9B_v4_half with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use HerrHruby/MR_midtrain_9B_v4_half with Transformers:

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

pipe = pipeline("text-generation", model="HerrHruby/MR_midtrain_9B_v4_half")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("HerrHruby/MR_midtrain_9B_v4_half")
model = AutoModelForCausalLM.from_pretrained("HerrHruby/MR_midtrain_9B_v4_half", 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 HerrHruby/MR_midtrain_9B_v4_half 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_half"
# 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_half",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/HerrHruby/MR_midtrain_9B_v4_half

SGLang

How to use HerrHruby/MR_midtrain_9B_v4_half 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_half" \
    --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_half",
		"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_half" \
        --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_half",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use HerrHruby/MR_midtrain_9B_v4_half with Docker Model Runner:
```
docker model run hf.co/HerrHruby/MR_midtrain_9B_v4_half
```

MR_midtrain_9B_v4_half / README.md

HerrHruby

v4 midtrain SFT data-scaling point

ef4115e verified 27 days ago

preview code

Raw History Blame Contribute Delete

1.89 kB

	---
	base_model: Qwen/Qwen3.5-9B
	library_name: transformers
	tags: [meta-reasoning, sft, data-scaling]
	---

	# MR_midtrain_9B_v4 — half corpus

	A data-scaling point for [`MR_midtrain_9B_v4`](https://huggingface.co/HerrHruby/MR_midtrain_9B_v4):
	the same v4 midtrain SFT recipe on 37,404 rows / 2,244 problems, which is
	the half of the full corpus
	([`MR_midtrain_V4_sft`](https://huggingface.co/datasets/HerrHruby/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. Stopped at step 1600 of 1752 (91.3% through the cosine schedule, lr ~5.7e-7 vs a final 2.0e-7) because val loss had been flat to four decimals for 300 steps and the remaining allocation was better spent on the quarter run.

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