datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
YuLan-Mini-Text-Datasets
News
[2025.04.11] Add dataset mixture: link.
[2025.03.30] Text datasets upload finished.
This is text dataset.
这是文本格式的数据集。
Since we have used BPE-Dropout, in order to ensure accuracy, you can find the tokenized dataset here.
由于我们使用了BPE-Dropout，为了保证准确性，你可以在这里找到分词后的数据。
For more information, please refer to our datasets details and preprocess details.
Contributing
We welcome any form of contribution, including feedback on model bad cases, feature suggestions, and example… See the full description on the dataset page: https://huggingface.co/datasets/yulan-team/YuLan-Mini-Text-Datasets.MiniMax-M2.1-Mixture-of-Thoughts
MiniMax-M2.1 Mixture of Thoughts
This dataset contains responses generated by MiniMax-M2.1 for user questions from the open-r1/Mixture-of-Thoughts dataset.
Dataset Description
The dataset captures both the extended thinking process and final answers from MiniMax-M2.1, with reasoning wrapped in <think> tags for easy separation.
Metric
Value
Examples
349,317
Total Tokens
4,052,592,552
Avg Tokens/Example
11,601
Source Dataset
Name:… See the full description on the dataset page: https://huggingface.co/datasets/PursuitOfDataScience/MiniMax-M2.1-Mixture-of-Thoughts.minimax-m3-claude-code-tracesThis dataset was generated using teich by TeichAI
Prepare these datasets for supervised fine-tuning in just a few lines of code — see the Conversion section below.
Minimax M3 Claude Code Traces
This directory contains raw agent trace files generated by teich.
All assistant responses were generated by minimax/minimax-m3.
JSONL files: 31
Format
Each file is newline-delimited JSON representing a single captured agent session.
The trace schema is designed for… See the full description on the dataset page: https://huggingface.co/datasets/armand0e/minimax-m3-claude-code-traces.the-stack-mini-nonshuffledThis repo contains (up to) 30k samples of 21 languages (top 20 languages by StackOverflow survey, html/css was splited).
'javascript', 'html', 'css', 'python', 'sql', 'typescript', 'shell', 'java', 'c-sharp', 'cpp', 'c', 'php', 'powershell', 'go', 'rust', 'kotlin', 'lua', 'dart', 'assembly', 'ruby', 'swift'
role-play-bench
Role-play Benchmark
A comprehensive benchmark for evaluating Role-play Agents in Chinese and English scenarios.
Dataset Summary
Role-play Benchmark is designed to evaluate Role-play Agents' ability to deliver immersive role-play experiences through Situated Reenactment. Unlike traditional benchmarks with verifiable answers, Role-play is fundamentally non-verifiable, e.g., there's no single "correct" response when a tsundere character is asked "Do you like me?".… See the full description on the dataset page: https://huggingface.co/datasets/MiniMaxAI/role-play-bench.the-stack-minipoc-mini-trade-game-dataset
Dataset Card for Mini Trade Game NPC Dataset
Dataset Summary
This dataset contains synthetic training examples for simulating NPC (Non-Player Character) merchant behavior in a trading game scenario. The dataset is designed to train language models to generate contextually appropriate trading responses based on item properties, relationship status, and player interactions.
All examples are in Traditional Chinese (zh-TW), with player inputs and NPC responses using… See the full description on the dataset page: https://huggingface.co/datasets/aotoki/poc-mini-trade-game-dataset.minimax-m3-tiny-cpu-repro-v1
minimax-m3 complete native tiny random CPU fixture
Complete untrained MiniMaxM3SparseForConditionalGeneration checkpoint with an untied full LM head,
a real 272-entry byte tokenizer and every native state tensor. Architecture lineage:
MiniMaxAI/MiniMax-M3@f0e1c1e04d40177e4673a22097036854f536e9c0.
No upstream weights, training data, paid GPU or cloud compute were used.
Complete native image/text wrapper with real shrunk Conv3D vision, nonempty 3D RoPE, patch-merge projector and… See the full description on the dataset page: https://huggingface.co/datasets/malaiwah/minimax-m3-tiny-cpu-repro-v1.minimax-m2-tiny-cpu-repro-v1
minimax-m2 complete native tiny random CPU fixture
Complete untrained MiniMaxM2ForCausalLM checkpoint with an untied full LM head,
a real 272-entry byte tokenizer and every native state tensor. Architecture lineage:
MiniMaxAI/MiniMax-M2.7@d494266a4affc0d2995ba1fa35c8481cbd84294b.
No upstream weights, training data, paid GPU or cloud compute were used.
Complete native text causal LM: sigmoid/top-k MoE routing with correction bias, per-layer flattened Q/K RMSNorm and half-head… See the full description on the dataset page: https://huggingface.co/datasets/malaiwah/minimax-m2-tiny-cpu-repro-v1.MiniMax-M3-150k-Mixed
m3-alldomains-verified-107k
Verified distillation traces generated with faststill v0.0.1 — a pipeline that generates (prompt, reasoning, output) triplets from any OpenAI-compatible chat-completions endpoint and deterministically verifies every row before keeping it. A row is verified=true only when a machine check (executed unit tests, exact / normalized answer compare) confirmed it, so wrong labels are filtered out instead of poisoning a student model.
Dataset… See the full description on the dataset page: https://huggingface.co/datasets/empero-ai/MiniMax-M3-150k-Mixed.code-minier
Code minier, non-instruct (2025-09-18)
The objective of this project is to provide researchers, professionals and law students with simplified, up-to-date access to all French legal texts, enriched with a wealth of data to facilitate their integration into Community and European projects.
Normally, the data is refreshed daily on all legal codes, and aims to simplify the production of training sets and labeling pipelines for the development of free, open-source language models based… See the full description on the dataset page: https://huggingface.co/datasets/louisbrulenaudet/code-minier.mini-fineweb-edu
Mini FineWeb-Edu (CC-MAIN-2025-26)
Mini FineWeb-Edu (CC-MAIN-2025-26) is a lightweight subset of the original FineWeb-Edu dataset. To support different development and research workflows, the dataset is released in five progressively larger subsets: debug, tiny, small, base, and full.
The dataset is designed to provide a compact and practical corpus for developing and evaluating small language models (SLMs) and language model training pipelines, while significantly reducing… See the full description on the dataset page: https://huggingface.co/datasets/BaoMinhTran/mini-fineweb-edu.swebench-minimax-traces
swebench-minimax-traces
Complete HTTP-level agentic traces from running swebench benchmark tasks through an instrumented reverse proxy.
Each trace captures full request/response pairs including system prompts, user messages, assistant responses, tool calls and results, and token usage metadata.
Stats
Total sessions: 500
Multi-turn sessions (2+ LLM calls): 357
Total records: 22424
Total LLM requests: 11249
Format
Raw JSONL traces from the instrumented proxy.… See the full description on the dataset page: https://huggingface.co/datasets/sammshen/swebench-minimax-traces.minipile-graded
pszemraj/minipile-graded
eduscore added - a test. This model was used to classify/assign an eduscore to JeanKaddour/minipile
intercode-minimax-traces
intercode-minimax-traces
Complete HTTP-level agentic traces from running intercode benchmark tasks through an instrumented reverse proxy.
Each trace captures full request/response pairs including system prompts, user messages, assistant responses, tool calls and results, and token usage metadata.
Stats
Total sessions: 338
Multi-turn sessions (2+ LLM calls): 299
Total records: 5838
Total LLM requests: 2919
Format
Raw JSONL traces from the instrumented proxy.… See the full description on the dataset page: https://huggingface.co/datasets/sammshen/intercode-minimax-traces.minimax-m3-150k-mixed
m3-alldomains-verified-107k
Verified distillation traces generated with faststill v0.0.1 — a pipeline that generates (prompt, reasoning, output) triplets from any OpenAI-compatible chat-completions endpoint and deterministically verifies every row before keeping it. A row is verified=true only when a machine check (executed unit tests, exact / normalized answer compare) confirmed it, so wrong labels are filtered out instead of poisoning a student model.
Dataset… See the full description on the dataset page: https://huggingface.co/datasets/ansulev/minimax-m3-150k-mixed.minimax-m2.1-reap-observations
[!TIP]
Support this work: donate.sybilsolutions.ai
REAP surfaces: GLM | MiniMax | Qwen | Gemma | Paper | Code | PR17 | Cerebras Collection
MiniMax-M2.1 REAP Stress Test Observations
Comprehensive stress test results for MiniMax-M2.1 models pruned with REAP (Router-weighted Expert Activation Pruning) at various compression ratios.
Dataset Description
This dataset contains 96 stress test results across 4 pruned MiniMax-M2.1 models, testing for repetition loops at… See the full description on the dataset page: https://huggingface.co/datasets/0xSero/minimax-m2.1-reap-observations.qwen9b-coop-mini-swe-agent
qwen9b-coop-mini-swe-agent
Two-agent cooperative coding trajectories generated by running
CooperBench in coop mode on
the CooperData task set, using
Qwen/Qwen3.5-9B as the model and mini_swe_agent_v2 as the agent framework.
Each pair runs two agents in parallel — one per feature — coordinating via Redis messaging and a shared git remote.
The matched solo version is at
CooperBench/qwen9b-solo-mini-swe-agent.
Same task corpus, same model, same agent — only the coordination differs… See the full description on the dataset page: https://huggingface.co/datasets/CooperBench/qwen9b-coop-mini-swe-agent.GPT-OSS-20B-Distilled-Reasoning-Mini
Dataset Card for Dataset Name
GPT-OSS-20B Distilled Reasoning Dataset Mini
(Multi-stage Evaluative Refinement Method for Reasoning Generation)
Dataset Details and Description
This is a high-quality instruction fine-tuning dataset constructed through knowledge distillation, featuring detailed Chain-of-Thought (CoT) reasoning processes. The dataset is designed to enhance the capabilities of smaller language models in complex reasoning, logical analysis, and instruction… See the full description on the dataset page: https://huggingface.co/datasets/Jackrong/GPT-OSS-20B-Distilled-Reasoning-Mini.GrandMaster-PRO-MINI-RU
AI Platforms GrandMaster PRO MINI RU
Компактный русскоязычный instruction/chat датасет для SFT небольших локальных языковых моделей.
Датасет собран как практический обучающий микс для моделей, которым нужно лучше держать русский диалоговый стиль, отвечать на короткие пользовательские вопросы и работать в локальных inference-сценариях.
Состав
Репозиторий содержит Parquet-файлы:
openhermes_translated.parquet — переведенные/адаптированные instruction-примеры в стиле… See the full description on the dataset page: https://huggingface.co/datasets/aiplatforms/GrandMaster-PRO-MINI-RU.mint-minimax-traces
mint-minimax-traces
Complete HTTP-level agentic traces from running mint benchmark tasks through an instrumented reverse proxy.
Each trace captures full request/response pairs including system prompts, user messages, assistant responses, tool calls and results, and token usage metadata.
Stats
Total sessions: 242
Multi-turn sessions (2+ LLM calls): 145
Total records: 1388
Total LLM requests: 694
Format
Raw JSONL traces from the instrumented proxy. Each line is… See the full description on the dataset page: https://huggingface.co/datasets/sammshen/mint-minimax-traces.qwen9b-solo-mini-swe-agent
qwen9b-solo-mini-swe-agent
Single-agent coding trajectories generated by running
CooperBench in solo mode on
the CooperData task set, using
Qwen/Qwen3.5-9B as the model and mini_swe_agent_v2 as the agent framework.
One agent implements both features in each task.
The matched coop version is at
CooperBench/qwen9b-coop-mini-swe-agent.
Same task corpus, same model, same agent — only the coordination differs, so
together they isolate the cooperation deficit.
At a glance… See the full description on the dataset page: https://huggingface.co/datasets/CooperBench/qwen9b-solo-mini-swe-agent.swebench-minimax-traces-2
swebench-minimax-traces-2
Complete HTTP-level agentic traces from running swebench benchmark tasks through an instrumented reverse proxy.
Each trace captures full request/response pairs including system prompts, user messages, assistant responses, tool calls and results, and token usage metadata.
Stats
Total sessions: 250
Multi-turn sessions (2+ LLM calls): 210
Total records: 15198
Total LLM requests: 7623
Format
Raw JSONL traces from the instrumented proxy.… See the full description on the dataset page: https://huggingface.co/datasets/sammshen/swebench-minimax-traces-2.github-mini
github-mini
A dataset consisting of source code from GitHub repositories.
Dataset Curation
Source: GitHub repositories with 500 to 99,999+ stars.
Licenses: Filtered for MIT and Apache-2.0.
Focus: Source code across multiple programming languages.
Dataset Structure
Each record in the dataset contains the following fields:
repo_full_name: The full name of the repository (owner/name).
repo_url: Direct link to the GitHub repository.
stars: Number of stars at the… See the full description on the dataset page: https://huggingface.co/datasets/lumasik/github-mini.orca-mini-short-100tok
Orca-Mini - Short Context (<100 tokens)
This dataset is a filtered version of ermiaazarkhalili/orca-mini-v1-high-prob-qwen-0.5b-10k
containing only samples with fewer than 100 tokens.
Dataset Description
This dataset is designed for efficient XAI (Explainable AI) attribution evaluation on decoder models.
Short context samples allow for faster evaluation while maintaining meaningful attribution analysis.
Statistics
Total samples: 5,000
Token count range: 5 - 99… See the full description on the dataset page: https://huggingface.co/datasets/ermiaazarkhalili/orca-mini-short-100tok.fineweb-CC-MAIN-2024-10-insurance-700k-dedup-minifiedbased on https://hf.co/datasets/gogo8232/fineweb-CC-MAIN-2024-10-insurance-700k-dedup
cultivar-mini
CULTIVAR-Mini
Preview subset of the CULTIVAR benchmark dataset.
One record per split (train, val, test) — intended for schema exploration,
tooling development, and pipeline testing.
The full dataset (cultivar) is not yet publicly available.
Sub-dataset: kropp2026-maz
Full identifier: CULTIVAR-J-Kropp2026-111359-MAZ
Source: Kropp et al. (2026), Integrating interactive decision making into
evolutionary multiobjective agricultural optimization, Computers and Electronics
in… See the full description on the dataset page: https://huggingface.co/datasets/eusojk/cultivar-mini.controlled-generated-convos-gpt-4.1-mini
Controlled Generated Conversations: gpt-4.1-mini
Dataset Description
This dataset contains synthetic customer support conversations generated using gpt-4.1-mini as part of research on cross-lingual stability of LLM judges. The conversations are designed for evaluating how well language models maintain consistent performance across different languages, with a focus on Finno-Ugric languages (Estonian, Finnish, Hungarian) and English.
Dataset Summary
Languages:… See the full description on the dataset page: https://huggingface.co/datasets/isaacchung/controlled-generated-convos-gpt-4.1-mini.mini-enedina-dataset
Mini-Enedina Dataset: Physically Validated Timoshenko Shaft Analysis (60k)
Training dataset for Mini-Enedina 37.5M -- a monotropic language model deliberately small and intensively specialized for structural shaft analysis according to Timoshenko beam theory.
Dataset Description
60,000 synthetic conversations in Harmony-Enedina format (a ChatML variant), covering three progressively complex levels of shaft analysis:
Level
Analysis Scope
Samples
Avg. Tokens/Sample… See the full description on the dataset page: https://huggingface.co/datasets/aiacontext/mini-enedina-dataset.code-minier-nouveau
Code minier (nouveau), non-instruct (2025-09-11)
The objective of this project is to provide researchers, professionals and law students with simplified, up-to-date access to all French legal texts, enriched with a wealth of data to facilitate their integration into Community and European projects.
Normally, the data is refreshed daily on all legal codes, and aims to simplify the production of training sets and labeling pipelines for the development of free, open-source language… See the full description on the dataset page: https://huggingface.co/datasets/louisbrulenaudet/code-minier-nouveau.