# Base (no inference backend)
pip install neural-txt

# With HuggingFace backend (torch)
pip install neural-txt[hf]

# With MLX backend (Apple Silicon)
pip install neural-txt[mlx]

Quick start

python

from neuraltxt import NeuralTxt

model = NeuralTxt(backend="mlx")  # or backend="hf"

passage = """
Transformers have revolutionized NLP by introducing the self-attention
mechanism. Unlike RNNs, transformers process all tokens in parallel,
leading to significant training speedups.
"""

# Extract key points
bullets = model.extract_bullets(passage)

# Generate question-answer pairs
pairs = model.generate_qa_pairs(passage)

# Extract knowledge graph triplets
triplets = model.extract_triplets(passage)

JSON mode

Every method supports json=True for guaranteed structured output via outlines:

python

# Returns a BulletsOutput pydantic model
bullets = model.extract_bullets(passage, json=True)
print(bullets.bullets)  # list[str]

# Returns a QAPairsOutput pydantic model
qa = model.generate_qa_pairs(passage, json=True)
for pair in qa.pairs:
    print(pair.question, pair.answer)

# Returns a TripletsOutput pydantic model
triplets = model.extract_triplets(passage, json=True)
for t in triplets.triplets:
    print(t.subject, t.relation, t.object)

API

Method	Input	Output	JSON Output
`extract_bullets(passage)`	passage	`list[str]`	`BulletsOutput`
`generate_qa_pairs(passage)`	passage	`list[QAPair]`	`QAPairsOutput`
`generate_question(passage)`	passage	`str`	`QuestionOutput`
`generate_questions_list(passage)`	passage	`list[str]`	`QuestionsListOutput`
`extract_fact(passage)`	passage	`str`	`FactOutput`
`answer(question, passage)`	question + passage	`str`	`AnswerOutput`
`rephrase(passage)`	passage	`str`	`RephraseOutput`
`continue_from(passage)`	passage start	`str`	`ContinuationOutput`
`extract_triplets(passage)`	passage	`list[Triplet]`	`TripletsOutput`
`compare(passage_a, passage_b)`	two passages	`str`	`ComparisonOutput`
`find_relevant(question, passages)`	question + passage list	`RetrievalResult`	`RetrievalOutput`

Benchmarks

On research paper QA and retrieval tasks, this model handsomely beats other edge LMs.

Model	Overall	Faithful.	Correct.	Relev.	Complete
neuraltxt (135M)	3.52	3.98	3.12	3.88	3.12
Qwen3.5-0.8B (4-bit)	3.35	3.75	2.98	3.85	2.81
Qwen3.5-4B (4-bit)	3.31	3.99	3.16	3.24	2.84
Qwen3-0.6B	3.31	3.64	3.04	3.74	2.83
SmolLM2-135M-Instruct	2.38	2.73	2.15	2.63	1.99

How these models are made

All this is part of this YouTube course by Neural Breakdown:

Course Video 1 (CPT): https://youtu.be/B8Ur62D3J3U
Course Video 2 (SFT): https://youtu.be/gvZIUEL6Ruc
Course Video 3 (DPO): https://youtu.be/qQwg2nFtZ1Y

paperbd/neuraltxt-v1-135M

Install

Quick start

JSON mode

API

Benchmarks

How these models are made

paperbd/
neuraltxt-v1-135M