datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
pybytecode-mbpp-3.12
Data card — MBPP held-out decompilation benchmark
383 MBPP reference solutions compiled to Python 3.12 bytecode and paired with their source, with
the original task_id restored on every row. Built 2026-08-04 by tools/build_mbpp_ood.py.
Redistributable under CC-BY-4.0, provided NOTICES.md ships alongside.
The set
Rows
383 (400 candidates − 17 contaminated)
Source
google-research-datasets/mbpp, config full
Licence
CC-BY-4.0
task_id restored
383 /… See the full description on the dataset page: https://huggingface.co/datasets/BlazingCustoms/pybytecode-mbpp-3.12.mbpp-tr
MBPP-TR
MBPP (Mostly Basic Python Problems)
veri setinin Türkçe çevirisi. Orijinal veri setindeki gibi iki config içerir: sanitized ve full.
Her satırda görev açıklamasının İngilizce aslı ve Türkçe çevirisi birlikte yer alır; kod ve testler
orijinal veri setinden değiştirilmeden aktarılmıştır.
A Turkish translation of MBPP with both the sanitized and full configs. Each row contains the
original English task description and its Turkish translation; code and tests are copied… See the full description on the dataset page: https://huggingface.co/datasets/firatmio/mbpp-tr.mbppMBPP dataset annotated with ground-truth programming solutions, to enable evaluations for retrieval and retrieval-augmented code generation.
Please refer to code-rag-bench for more details.
ML4SE23_G1_MBPP-SCoT
ML4SE23_G1_MBPP-SCoT
MBPP enhanced dataset with Structured-Chain-of-Thought
ML4SE23_G1_MBPP-SCoTMBPP enhanced dataset with Structured-Chain-of-Thought
MBPP-Thinking-Gate-1k
MBPP Thinking-Gate SFT Dataset
This package contains two related assets:
Ready 1,000-row MBPP-style dataset (all.jsonl, train.jsonl, validation.jsonl).
It is synthetic and designed to test/train autonomous routing between <DIRECT> and <THINK>.
Official-MBPP builder (build_from_official_mbpp.py).
Run this to create the production dataset from the official Google Research MBPP source.
Why two response modes?
The training target starts with one of two routing… See the full description on the dataset page: https://huggingface.co/datasets/islam-kamel/MBPP-Thinking-Gate-1k.burmese-mbpp
Burmese MBPP: A Large-Scale Programming Dataset for Burmese Coding Assistants
Dataset Summary
The Burmese MBPP dataset is a translated and augmented version of the Google Mostly Basic Python Problems (MBPP) benchmark. It is designed to facilitate the training and evaluation of Large Language Models (LLMs) in generating Python code from Burmese natural language instructions.
This dataset contains 974 programming tasks, each featuring:
Burmese Instructions: Formal and… See the full description on the dataset page: https://huggingface.co/datasets/WYNN747/burmese-mbpp.fuzzeval-humaneval-mbpp
FuzzEval unit tests for HumanEval-f and MBPP-f
Automatically generated unit tests for a reproduction of the ICML 2026 paper
"Towards Functional Correctness of Large Code Models with Selective Generation"
(Jeong, Kim & Park — arXiv:2505.13553,
official repo trustml-lab/selective-code-generation).
The paper's FuzzEval paradigm replaces a benchmark's handful of hand-written
unit tests with hundreds of unit tests obtained by fuzzing the reference
solution. This dataset is our… See the full description on the dataset page: https://huggingface.co/datasets/ababa134/fuzzeval-humaneval-mbpp.MBPP-Bangla
🐯 MBPP-Bangla: A Benchmark for Evaluating Bangla Code Generation
Accepted at LREC 2026
Nishat Raihan, Antonios Anastasopoulos, Marcos Zampieri
George Mason University, Fairfax, VA, USA
The first expert-validated, multi-language Bangla code generation benchmark with 974 problems across 5 programming languages.
⚠️ Note: The benchmark will be released after the LREC 2026 conference. Stay tuned!
Overview
MBPP-Bangla is a… See the full description on the dataset page: https://huggingface.co/datasets/md-nishat-008/MBPP-Bangla.