Skip to content
Dataset By Humans
On requestReasoning· AI reasoning

Vietnamese Math Reasoning Dataset: Step-by-Step Solutions

Secondary-school and university-entrance math problems in Vietnamese, each solved step by step by a qualified teacher and checked by a second one.

On request: this dataset is produced to your specification after you order. Typically 2–3 weeks for 300 examples and 5–7 weeks for 1,500.

Pricing

Fixed prices for standard packages, produced to your spec. Pick a package and a license to see the exact total, or ask for a custom quote.

Packages compared
FeatureStarterProCustom
examples3001,500Any quantity
What's included
  • 300 examples in JSONL
  • Topic, level and difficulty labels
  • Datasheet, schema, license and checksums
  • 1,500 examples in JSONL
  • Everything in Starter
  • Step-level correctness labels
  • Topic mix agreed with you
Your quantity, annotations and deadline
Datasheet, license, checksumsIncludedIncludedIncluded
Research license$490$1.63 / example$1,890$1.26 / exampleQuote
Commercial license$1,490$4.97 / example$5,900$3.93 / exampleQuote
Exclusive license$4,900$16.33 / example$18,900$12.60 / exampleQuote

How ordering works

  1. 1Choose a package and license and place your order. No payment is taken online.
  2. 2Within 48 hours our team contacts you to confirm the spec and invoices a 50% deposit.
  3. 3We produce and deliver in parts so you can review early. Typically 2–3 weeks for 300 examples and 5–7 weeks for 1,500.
  4. 4Pay the balance and download the final dataset.

Custom quote

Need another quantity, an Enterprise license or annotations for this dataset? Available here:

  • Difficulty and topic labels
  • Step-level correctness labels
  • Common-mistake (negative) solutions
  • English translation
Request a custom quote

Overview

Large language models are increasingly judged on how they reason, not only on the final answer. Public math reasoning sets are overwhelmingly English and often generated by other models, so errors and stylistic tics get copied from one model to the next.

This dataset is written by Vietnamese math teachers. Each record holds a problem in natural Vietnamese, a numbered chain of reasoning steps the way a good teacher would explain it, and the final answer. A second teacher works through every solution independently before it is accepted.

Example records

Illustrative records in the delivered format. Request the sample pack for real records written by our experts.

  • Record 1 of 3 · illustrative

    {
      "id": "vmr-000214",
      "level": "grade-12",
      "steps": [
        "Điều kiện xác định: x + 3 > 0 và x − 1 > 0, suy ra x > 1.",
        "Áp dụng log₂a + log₂b = log₂(ab): log₂[(x + 3)(x − 1)] = 5.",
        "Suy ra (x + 3)(x − 1) = 2⁵ = 32.",
        "Khai triển: x² + 2x − 3 = 32, tức x² + 2x − 35 = 0.",
        "Phân tích: (x + 7)(x − 5) = 0, nên x = 5 hoặc x = −7.",
        "Đối chiếu điều kiện x > 1: chỉ nhận x = 5."
      ],
      "topic": "logarithms",
      "problem": "Giải phương trình log₂(x + 3) + log₂(x − 1) = 5.",
      "language": "vi",
      "difficulty": "medium",
      "author_role": "high-school math teacher",
      "final_answer": "x = 5",
      "verified_by_second_teacher": true
    }
  • Record 2 of 3 · illustrative

    {
      "level": "grade-7",
      "steps": [
        "Gọi giá ban đầu là x (đồng).",
        "Sau lần giảm 20%, giá còn 0,8x.",
        "Sau lần giảm thêm 10%, giá còn 0,9 × 0,8x = 0,72x.",
        "Ta có 0,72x = 360.000, nên x = 360.000 : 0,72 = 500.000."
      ],
      "problem": "Một cửa hàng giảm giá 20% cho một chiếc áo, sau đó giảm thêm 10% trên giá đã giảm. Giá cuối cùng là 360.000 đồng. Hỏi giá ban đầu của chiếc áo?",
      "final_answer": "500.000 đồng"
    }
  • Record 3 of 3 · illustrative

    {
      "level": "grade-12",
      "steps": [
        "Điều kiện x > 1.",
        "(x + 3)(x − 1) = 32.",
        "x² + 2x − 35 = 0 ⇒ x = 5 hoặc x = −7.",
        "Đối chiếu điều kiện: x = 5."
      ],
      "problem": "Giải phương trình log₂(x + 3) + log₂(x − 1) = 5.",
      "final_answer": "x = 5"
    }

Get a free sample pack

We email you real examples from this dataset.

Used only to send your samples. Privacy policy.

Technical specifications

Authors
Qualified math teachers (secondary and university-entrance level)
Language
Vietnamese; English translations on request
Levels
Grades 6–12 and university-entrance exam level
Topics
Arithmetic, algebra, geometry, probability, logarithms, calculus, word problems
Review
Every solution re-solved independently by a second teacher
Format
JSONL (one record per line), UTF-8; CSV on request
AI use
Written by people; screened for LLM-generated text. Any AI assistance is disclosed in the datasheet
License
Research, Commercial, Enterprise or Exclusive

Use cases

  • Reasoning fine-tuning

    Supervised fine-tuning on human-written chains of thought in Vietnamese, from arithmetic to calculus.

  • Reward and process models

    Step-level correctness labels let you train process reward models that judge each step, not only the answer.

  • Held-out evaluation

    Order an Exclusive set that never appears online and measure real progress without contamination.

  • EdTech tutors

    Teach tutoring assistants to explain solutions the way Vietnamese students are taught in class.

How this data is made

  1. 1Problems sourced or written by vetted domain experts
  2. 2Each solution written step by step by a person, never generated
  3. 3A second expert checks every step and the final answer
  4. 4Disagreements resolved by a lead reviewer
  5. 5Screened for copied text and LLM-generated writing
  6. 6Difficulty, topic and skill labels added
  7. 7Exported as JSONL with a schema you can rely on
  8. 8Datasheet and license packaged with the delivery
Read our full process

Provenance & legal

  • 100% made by people: no scraping, no generative AI
  • Datasheet documenting how the data was made, checked and its limitations
  • Commercial license that lets you keep models trained on the data
  • Supports training-data documentation under the EU AI Act
Licensing options

Dataset-specific notes

  • Problems are written fresh or adapted with permission; no copying from textbooks or exam archives under copyright.
  • Each step is a separate string, so you can mask, score or truncate steps individually.
  • Wrong-but-plausible solutions with the error marked are available for preference and reward-model training.

Delivery format

One JSON record per line, with the fields shown in the example records above and a SCHEMA.md describing each one.

Each delivery contains:

vietnamese-math-reasoning-v1/
├── data/train.jsonl
├── data/sample.jsonl
├── SCHEMA.md
├── DATASHEET.md
├── LICENSE.pdf
└── checksums.sha256

Frequently asked questions

Are the solutions generated by an AI and then edited?

No. Solutions are written by teachers from scratch. We screen submissions for LLM-generated text, and the datasheet states that no generative AI was used to write them.

How do you make sure answers are correct?

A second teacher solves every problem independently. If the answers or key steps disagree, a lead reviewer decides and the record is fixed or dropped.

Can you match our curriculum or exam format?

Yes. Send the syllabus, exam format or example problems and we will propose a topic mix in the quote.

How does ordering and payment work?

Create a free account and place your order on this page; nothing is charged online. Our team contacts you to confirm the spec and invoices a 50% deposit, with the balance due on final delivery. The deposit is refunded in full if we cannot deliver the agreed spec.

What is the difference between non-exclusive and Exclusive?

With Research or Commercial we may license the examples produced for your order to other buyers later. With Exclusive they are never licensed to anyone else or added to our catalog.

Can I order a different quantity or extra annotations?

Yes. Request a custom quote with the quantity, annotations and deadline you need. Prices per record fall as quantity grows.

Need a variation?

Different topics, languages, difficulty, quantity or labels? Describe it and we reply within 48 hours with matching samples and a quote.

Request a quote
Browse all datasets