Instructions to use lyrain2001/Auto-Fill-Qwen3-8B-Knowledge with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lyrain2001/Auto-Fill-Qwen3-8B-Knowledge with Transformers:

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

pipe = pipeline("text-generation", model="lyrain2001/Auto-Fill-Qwen3-8B-Knowledge")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("lyrain2001/Auto-Fill-Qwen3-8B-Knowledge")
model = AutoModelForCausalLM.from_pretrained("lyrain2001/Auto-Fill-Qwen3-8B-Knowledge", 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 lyrain2001/Auto-Fill-Qwen3-8B-Knowledge with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lyrain2001/Auto-Fill-Qwen3-8B-Knowledge"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lyrain2001/Auto-Fill-Qwen3-8B-Knowledge",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/lyrain2001/Auto-Fill-Qwen3-8B-Knowledge

SGLang

How to use lyrain2001/Auto-Fill-Qwen3-8B-Knowledge 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 "lyrain2001/Auto-Fill-Qwen3-8B-Knowledge" \
    --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": "lyrain2001/Auto-Fill-Qwen3-8B-Knowledge",
		"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 "lyrain2001/Auto-Fill-Qwen3-8B-Knowledge" \
        --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": "lyrain2001/Auto-Fill-Qwen3-8B-Knowledge",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use lyrain2001/Auto-Fill-Qwen3-8B-Knowledge with Docker Model Runner:
```
docker model run hf.co/lyrain2001/Auto-Fill-Qwen3-8B-Knowledge
```

Auto-Fill Knowledge Specialist (Qwen3-8B)

The knowledge specialist of Auto-Fill: Learning to Predict Missing Values Accurately with Specialist Language Models (PVLDB 19(11), 2026 — arXiv:2607.19847). Given a table with one cell marked [MISSING], this model answers directly, without chain-of-thought, and is meant for cells whose value is a matter of world knowledge or of patterns visible in the table (entity attributes, codes, names, dates).

Auto-Fill runs three specialists in parallel — knowledge, reasoning and coding — calibrates their confidences with isotonic regression and returns the most confident answer (or abstains). Sibling specialists: Reasoning · Coding.

Code: https://github.com/lyrain2001/auto-fill
Benchmark: lyrain2001/Auto-Fill-Benchmark
Calibrators: calibrators.json in this repo (also in the code repository under checkpoints/)

Model details


Base model	Qwen/Qwen3-8B
Training	full-parameter SFT with ms-swift, DeepSpeed ZeRO-3, bf16
Data	30,000 examples: direct `(table, value)` pairs — one cell masked per table, the original value as target.
Hyper-parameters	2 epochs, lr 1e-5, cosine schedule, weight decay 0.1, max length 40,960, effective batch size 16
Hardware	4× A100 80 GB
Confidence signal	the log-probability of the generated value tokens (obtained from vLLM `logprobs`). Calibrated with isotonic regression fitted on a held-out validation split (`calibrators.json`).
Decoding used in the paper	temperature 0.1, max new tokens 4096, default Qwen3 chat template

Training tables come from public sources only (spreadsheets crawled from a search-engine index, public BI models, Wikipedia, nationalarchives.gov.uk, GitHub CSV/Parquet files); one cell per table is masked and its original value is the target.

Prompt and output format

The table is serialized as a Markdown pipe table (pandas.DataFrame.to_markdown(index=False, tablefmt="pipe")) with the cell to fill written as [MISSING]. The user message is exactly (see autofill/utils/prompts.py):

Please fill in the missing value in the input table. The missing value is denoted by '[MISSING]'. Please return the value filled in JSON format: {"value": "filled_value"}.

Input Table:
<markdown table>

Expected output: {"value": "<filled value>"} — no reasoning text.

Usage

With the code repository (recommended) — runs the full ensemble on one table:

python inference/run_specialists.py \
    --table          /path/to/table.csv \
    --knowledge_path lyrain2001/Auto-Fill-Qwen3-8B-Knowledge \
    --reasoning_path lyrain2001/Auto-Fill-Qwen3-8B-Reasoning \
    --coding_path    lyrain2001/Auto-Fill-Qwen3-8B-Coding \
    --calibrators    checkpoints/calibrators.json \
    --gpu_ids        0,1,2

or this specialist alone on the benchmark:

python inference/run_benchmark.py --mode knowledge --model_path lyrain2001/Auto-Fill-Qwen3-8B-Knowledge \
    --dataset Gov-CSV --benchmark Auto-Fill-Benchmark/sample200 --gpu_ids 0

Minimal vLLM example

import pandas as pd
from vllm import LLM, SamplingParams

llm = LLM(model="lyrain2001/Auto-Fill-Qwen3-8B-Knowledge", dtype="bfloat16", max_model_len=40960)
table = pd.read_csv("table.csv", dtype=str).to_markdown(index=False, tablefmt="pipe", disable_numparse=True)
prompt = PROMPT + table   # PROMPT = the user message above, up to and including "Input Table:\n"
text = llm.get_tokenizer().apply_chat_template(
    [{"role": "user", "content": prompt}], tokenize=False, add_generation_prompt=True)
out = llm.generate([text], SamplingParams(temperature=0.1, max_tokens=4096))
print(out[0].outputs[0].text)

Results

Recall@Precision=0.9 on the Auto-Fill benchmark (200 cases per dataset; from the paper's specialist ablation):

	Pub-XLS	Pub-BI	Pub-Wiki	Gov-CSV	Git-Parquet	Ent-CSV*	Ent-XLS*	Pub-Web	Rel-AR	Rel-FD	Rel-ST	Mean
Knowledge specialist alone	0.335	0.580	0.210	0.450	0.565	0.570	0.640	0.235	0.550	0.865	0.960	0.542
Auto-Fill (all three + ensemble)	0.525	0.615	0.285	0.500	0.590	0.585	0.660	0.275	0.990	0.890	0.995	0.628

* Ent-CSV / Ent-XLS are proprietary enterprise datasets that are not part of the public benchmark.

Limitations

Trained and evaluated on English-language tables with one missing cell per table; tables were serialized with at most 40,960 tokens.
The model can be wrong with high confidence on cells that require knowledge outside the table; use the calibrated confidence and abstain below a threshold, as in the paper.
Generated code (coding specialist) should be executed in a sandbox.

Citation

@article{liu2026autofill,
  title={Auto-Fill: Learning to Predict Missing Values Accurately with Specialist Language Models},
  author={Liu, Yurong and He, Yeye and Dong, Haoyu and Xing, Junjie and Han, Shi and Zhang, Dongmei and Chaudhuri, Surajit},
  journal={Proceedings of the VLDB Endowment},
  volume={19},
  number={11},
  pages={3160--3173},
  year={2026}
}