Instructions to use CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16_olddata with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16_olddata with Transformers:

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

pipe = pipeline("text-generation", model="CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16_olddata")
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("CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16_olddata")
model = AutoModelForCausalLM.from_pretrained("CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16_olddata", 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 CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16_olddata with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16_olddata"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16_olddata",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16_olddata

SGLang

How to use CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16_olddata 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 "CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16_olddata" \
    --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": "CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16_olddata",
		"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 "CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16_olddata" \
        --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": "CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16_olddata",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16_olddata with Docker Model Runner:
```
docker model run hf.co/CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16_olddata
```

Qwen3-8B-Base-CPT-BF16_olddata / README.md

ganscs

Validated standalone BF16 epoch 2, eight shards

55ddd73 verified about 1 month ago

preview code

Raw History Blame Contribute Delete

3.07 kB

	---
	license: apache-2.0
	library_name: transformers
	pipeline_tag: text-generation
	language: en
	base_model: Qwen/Qwen3-8B-Base
	tags:
	- continued-pretraining
	- merged
	- bfloat16
	---

	# Qwen3-8B-Base-CPT-BF16

	Standalone BF16 (16-bit) continued-pretraining model. This revision contains
	the complete model from epoch 2, step 750, in exactly **eight
	safetensors weight shards**. Load directly with Transformers.

	## Epochs

	`main` contains epoch 2, the best training-time validation checkpoint. Each
	completed epoch is independently loadable using its `epoch-N` revision. Both
	training runs stopped after four epochs; there is no epoch-five export.

	\| Revision \| Training step \| Training-time validation loss \|
	\|---\|---:\|---:\|
	\| [epoch-1](https://huggingface.co/CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16/tree/epoch-1) \| 375 \| 1.2472849 \|
	\| [epoch-2](https://huggingface.co/CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16/tree/epoch-2) \| 750 \| 1.2268945 \|
	\| [epoch-3](https://huggingface.co/CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16/tree/epoch-3) \| 1125 \| 1.2514912 \|
	\| [epoch-4](https://huggingface.co/CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16/tree/epoch-4) \| 1500 \| 1.2871689 \|

	These losses were measured during adapter training, not by a new evaluation of
	the BF16 exports. Evaluate the four revisions separately to measure behavior
	changes across epochs.

	## Load a specific epoch

	```python
	import torch
	from transformers import AutoModelForCausalLM, AutoTokenizer

	model_id = "CompassioninMachineLearning/Qwen3-8B-Base-CPT-BF16"
	revision = "main"
	tokenizer = AutoTokenizer.from_pretrained(model_id, revision=revision)
	model = AutoModelForCausalLM.from_pretrained(
	model_id, revision=revision, dtype=torch.bfloat16, device_map="auto"
	)
	```

	## Merge provenance and validation

	- Original base: `Qwen/Qwen3-8B-Base` at `49e3418fbbbca6ecbdf9608b4d22e5a407081db4`.
	- Source adapter: [checkpoint 750](https://huggingface.co/ganscs/Qwen3-8b-qwen-h100-20260909-CPT-LoRA-checkpoints/tree/1586e45fc7c3f7d5e601dc454ba7e15884a2f1df/checkpoint-750).
	- Adapter repository revision: `1586e45fc7c3f7d5e601dc454ba7e15884a2f1df`.
	- Adapter SHA-256: `62933de6ac6e2484a5082afc7d4748c2267368cc7a3331fd874b93f4fe0e609b`.
	- Merged 252 rsLoRA layers with PEFT's safe merge.
	- The separately trained `embed_tokens` and `lm_head` matrices are included.

	Training used a 4-bit base; these exports merge the trained adapter into the
	pinned original BF16 base. The merge was performed using PEFT's safe merge,
	not Unsloth's export helper. Every tensor was checked for finite BF16 values and
	the complete architecture's names and shapes. Every epoch passed a standalone
	load, finite-logit forward pass and short greedy generation. No separate adapter
	is required, and optimizer/trainer checkpoint files are not included.

	`merge_manifest.json` records source and output checksums. `validation.json`
	records the inference smoke test. `training_manifest.json` preserves the training
	settings. Shard count controls packaging; BF16 specifies the 16-bit precision.