Instructions to use sfanm/d24-v6.2-0.5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use sfanm/d24-v6.2-0.5 with Transformers:

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

pipe = pipeline("text-generation", model="sfanm/d24-v6.2-0.5")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("sfanm/d24-v6.2-0.5")
model = AutoModelForCausalLM.from_pretrained("sfanm/d24-v6.2-0.5", 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 sfanm/d24-v6.2-0.5 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "sfanm/d24-v6.2-0.5"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sfanm/d24-v6.2-0.5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/sfanm/d24-v6.2-0.5

SGLang

How to use sfanm/d24-v6.2-0.5 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 "sfanm/d24-v6.2-0.5" \
    --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": "sfanm/d24-v6.2-0.5",
		"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 "sfanm/d24-v6.2-0.5" \
        --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": "sfanm/d24-v6.2-0.5",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use sfanm/d24-v6.2-0.5 with Docker Model Runner:
```
docker model run hf.co/sfanm/d24-v6.2-0.5
```

D24 v6.2-0.5

Public preservation release for the half branch of the D24 v6.2 ratio-20 experiment. This repository contains both the terminal midtraining state and its terminal SFT descendant.

Repository layout

Repository root: final SFT Transformers model (BF16, SimpleChatML).
midtrain-hf/: final midtraining Transformers model (BF16 base model).
megatron-midtrain/iter_0002200/: exact terminal Megatron checkpoint, including optimizer/training state.
megatron-midtrain/: trackers, completion receipt, and frozen config.
provenance/: frozen campaign tickets and SFT receipt/config.

from transformers import AutoModelForCausalLM, AutoTokenizer

repo = "sfanm/d24-v6.2-0.5"
sft_model = AutoModelForCausalLM.from_pretrained(repo, token=True)
sft_tokenizer = AutoTokenizer.from_pretrained(repo, token=True)

midtrain_model = AutoModelForCausalLM.from_pretrained(
    repo, subfolder="midtrain-hf", token=True
)

Lineage

Stage	Source/final state
Pretraining	iteration 44,000; 184,549,376,000 tokens
Midtraining	iteration 2,200; 9,227,468,800 tokens
SFT	iteration 1,773; 475,114,114 packed tokens

Historical midtraining used the experiment's branch-scaled WSD schedule: warmup_iters=92 and decay_iters=440 with cosine decay. SFT used a constant learning rate of 1e-4 from optimizer step zero with zero warmup.

Both portable exports contain 756,819,456 parameters in BF16. This repository is an artifact-preservation release with frozen lineage and integrity metadata.

Downloads last month: 16

Safetensors

Model size

0.8B params

Tensor type

BF16