datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
the-stack-smol
Dataset Description
A small subset (~0.1%) of the-stack dataset, each programming language has 10,000 random samples from the original dataset. The dataset has 2.6GB of text (code).
Languages
The dataset contains 30 programming languages:
"assembly", "batchfile", "c++", "c", "c-sharp", "cmake", "css", "dockerfile", "fortran", "go", "haskell", "html", "java",
"javascript", "julia", "lua", "makefile", "markdown", "perl", "php", "powershell", "python", "ruby", "rust"… See the full description on the dataset page: https://huggingface.co/datasets/bigcode/the-stack-smol.the-stack-v2-smollm3
The Stack v2 — materialized source code
Upstream dataset:
bigcode/the-stack-v2
Exact upstream commit:
e565caa3a78c2423bd374333a472b049eb090e47
Primary source-content endpoint:
https://softwareheritage.s3.amazonaws.com/content/{blob_id}
Configurations
TypeScript
Swift
Ruby
Rust
Go
Shell
Jupyter_Notebook
HTML
Python
Java
JavaScript
C
C++
C-Sharp
PHP
SQL
Markdown
Added columns
content: decoded source content
download_error: null on successful… See the full description on the dataset page: https://huggingface.co/datasets/jordangong/the-stack-v2-smollm3.smollm-corpus-cleaned
SmolLM-Corpus: Now shuffled and sharded (and Cleaned)!
This is a version of the SmolLM-Corpus where the 3 subsets have been interleved, shuffled and sharded as 23698 jsonl.zst files for easy streaming!
The dataset is comprised of the cosmopedia-v2 and fineweb-edu-dedup subsets from the original SmolLM-Corpus repo, with the python-edu subset being pulled from my python-edu-cleaned repo.
Dataset Structure
The dataset is split into 24 subdirectories, with the first 23… See the full description on the dataset page: https://huggingface.co/datasets/Avelina/smollm-corpus-cleaned.the-stack-smol-xl
Dataset Description
A small subset of the-stack dataset, with 87 programming languages, each has 10,000 random samples from the original dataset.
Languages
The dataset contains 87 programming languages:
'ada', 'agda', 'alloy', 'antlr', 'applescript', 'assembly', 'augeas', 'awk', 'batchfile', 'bison', 'bluespec', 'c',
'c++', 'c-sharp', 'clojure', 'cmake', 'coffeescript', 'common-lisp', 'css', 'cuda', 'dart', 'dockerfile', 'elixir',
'elm', 'emacs-lisp','erlang'… See the full description on the dataset page: https://huggingface.co/datasets/bigcode/the-stack-smol-xl.the-stack-smol-xs\smoltalk-chinese
Chinese SmolTalk Dataset [中文] [English]
[OpenCSG Community] [👾github] [wechat] [Twitter]
📖Technical Report
smoltalk-chinese is a Chinese fine-tuning dataset constructed with reference to the SmolTalk dataset. It aims to provide high-quality synthetic data support for training large language models (LLMs). The dataset consists entirely of synthetic data, comprising over 700,000 entries. It is specifically designed to enhance the performance of Chinese… See the full description on the dataset page: https://huggingface.co/datasets/opencsg/smoltalk-chinese.jupyter-scripts-smollm3
The Stack v2 Jupyter Notebooks as Scripts
This dataset contains script representations of the Jupyter notebooks in
The Stack v2. It was
created from the materialized Jupyter_Notebook split in
jordangong/the-stack-v2-smollm3.
The output schema follows the Jupyter-script schema used by
bigcode/starcoderdata,
but this release is not deduplicated, PII-filtered, or otherwise equivalent
to StarCoderData's filtered split.
Relationship to the SmolLM3 training mix
This… See the full description on the dataset page: https://huggingface.co/datasets/jordangong/jupyter-scripts-smollm3.smolmo-sft-v2-seqlen64k
smolmo-sft-v2-seqlen64k
A supervised fine-tuning (SFT) dataset of math problems with full chain-of-thought solutions,
formatted for the Olmo 3 "Thinking" models.
2,813,055 examples · ~37.9 B tokens.
Three task families: proofs, numeric-answer problems, and tool-augmented (Python) problems.
Every assistant turn carries an explicit <think> … </think> reasoning trace before the answer.
Olmo 3 native chat + function-calling format; every example fits within a 64k-token context.… See the full description on the dataset page: https://huggingface.co/datasets/chankhavu/smolmo-sft-v2-seqlen64k.apigen-smollm-trl-FC
Dataset card for argilla-warehouse/apigen-smollm-trl-FC
This dataset is a merge of argilla/Synth-APIGen-v0.1
and Salesforce/xlam-function-calling-60k, and was prepared for training using the script
prepare_for_sft.py that can be found in the repository files.
References
@article{liu2024apigen,
title={APIGen: Automated Pipeline for Generating Verifiable and Diverse Function-Calling Datasets},
author={Liu, Zuxin and Hoang, Thai and Zhang, Jianguo and Zhu, Ming and… See the full description on the dataset page: https://huggingface.co/datasets/argilla-warehouse/apigen-smollm-trl-FC.smolkalam-arabic-conversational-sft
SmolKalam
SmolKalam is a quality-filtered Arabic SFT dataset of 1,790,478 examples (~2.45B tokens), built as an ensemble translation of SmolTalk2. It covers multi-turn dialogue (23% of rows), reasoning traces (19% carry <think>), tool and function calling (4.4%), and long context, categories that are underrepresented in existing Arabic post-training data. The SmolTalk2 source mixtures are kept as subsets.
Released with the paper SmolKalam: Ensemble Quality-Filtered Translation… See the full description on the dataset page: https://huggingface.co/datasets/AdaMLLab/smolkalam-arabic-conversational-sft.the-stack-v2-train-smol-ids
The Stack v2
The dataset consists of 4 versions:
bigcode/the-stack-v2: the full "The Stack v2" dataset
bigcode/the-stack-v2-dedup: based on the bigcode/the-stack-v2 but further near-deduplicated
bigcode/the-stack-v2-train-full-ids: based on the bigcode/the-stack-v2-dedup dataset but further filtered with heuristics and spanning 600+ programming languages. The data is grouped into repositories.
bigcode/the-stack-v2-train-smol-ids: based on the bigcode/the-stack-v2-dedup… See the full description on the dataset page: https://huggingface.co/datasets/bigcode/the-stack-v2-train-smol-ids.smol-smoltalk-Interaction-SFT
Dataset Card for ReactiveAI/Smol-Smoltalk Interaction SFT
Derived from HuggingFaceTB/smol-smoltalk. Made for Interaction Supervised Fine-Tuning of Reactive Transformer
Proof-of-Concept models, especially RxT-Beta.
Dataset Details
Dataset Description
Reactive Transformers are processing only the single interactions in real-time and using Short-Term Memory to store information from previous interactions.
Before the model is able to use it's memory, it has to be… See the full description on the dataset page: https://huggingface.co/datasets/ReactiveAI/smol-smoltalk-Interaction-SFT.smoltalk-ko-translate
번역 결과에 토큰이 반복된 결과들이 포함되어 있습니다. 필터링 후 재업로드 하겠습니다.
Z 알고리즘을 사용해 결과를 필터링 하였으며 {subset}_filtered 로 업로드하였습니다.
필터링 후 결과
subset
전
후
split/train
4205413
4162254
split/test
221249
218830
merge/train
1043917
1034473
merge/test
54948
54430
HuggingFaceTB/smoltalk 데이터셋의 subset:all을 nayohan/llama3-instrucTrans-enko-8b 모델을 사용해 번역했습니다.
원본의 messages 중 4096 token 이 넘어가는 content가 있다면 해당 레코드는 번역하지 않았습니다.
smol-worldcup
🏟️ Smol AI WorldCup — SHIFT Benchmark
The world's first 5-axis evaluation framework for small language models.
Not just "how smart?" — but "how honest? how fast? how small? how efficient?"
🏟️ Leaderboard
huggingface.co/spaces/ginigen-ai/smol-worldcup
📊 Dataset
huggingface.co/datasets/ginigen-ai/smol-worldcup
🏅 ALL Bench
huggingface.co/spaces/FINAL-Bench/all-bench-leaderboard
🏆 Official Ranking: WCS (WorldCup Score)
WCS = √( SHIFT × PIR_norm )… See the full description on the dataset page: https://huggingface.co/datasets/ginigen-ai/smol-worldcup.smollm-corpus-fineweb-edu-enPurified-openai-messages
📖 smollm-corpus-fineweb-edu-enPurified-openai-messages
smollm-corpus-fineweb-edu-enPurified is a highly curated, "prose-first" subset of the fineweb-edu-dedup subset found in HuggingFaceTB/smollm-corpus.
The enPurified collection is built on a specific philosophy: Specialization. While the original dataset is excellent for general pre-training, high-quality fluent English prose often gets diluted when mixed with syntax-heavy code, rigid math formulas, or low-information web junk.… See the full description on the dataset page: https://huggingface.co/datasets/enPurified/smollm-corpus-fineweb-edu-enPurified-openai-messages.smollm-corpus-cosmopedia-v2-enPurified-openai-messages
enPurified Collection: Smollm Corpus Cosmopedia V2]
Updated on January 15th to remove more math, code, and low quality English. The dataset has now been pruned from 39.1M rows down to ~9M rows.
Purpose of the enPurified Collection
The enPurified dataset collection is an initiative to curate strict, high-quality English prose datasets for language modeling. While the open-source community provides extensive resources for code, mathematics, and multilingual data, this… See the full description on the dataset page: https://huggingface.co/datasets/enPurified/smollm-corpus-cosmopedia-v2-enPurified-openai-messages.smolified-fakejob
🤏 smolified-fakejob
Intelligence, Distilled.
This is a synthetic training corpus generated by the Smolify Foundry.
It was used to train the corresponding model Aioshi/smolified-fakejob.
📦 Asset Details
Origin: Smolify Foundry (Job ID: 7b496b14)
Records: 597
Type: Synthetic Instruction Tuning Data
⚖️ License & Ownership
This dataset is a sovereign asset owned by Aioshi.
Generated via Smolify.ai.
smoltalk-chinese
Chinese SmolTalk Dataset [中文] [English]
[OpenCSG Community] [👾github] [wechat] [Twitter]
📖Technical Report
smoltalk-chinese is a Chinese fine-tuning dataset constructed with reference to the SmolTalk dataset. It aims to provide high-quality synthetic data support for training large language models (LLMs). The dataset consists entirely of synthetic data, comprising over 700,000 entries. It is specifically designed to enhance the performance of Chinese… See the full description on the dataset page: https://huggingface.co/datasets/sunorme/smoltalk-chinese.smoltalk-chinese-QwQ-Distrill
smoltalk-chinese-QwQ-Distrill [中文] [English]
📖Technical Report
smoltalk-chinese-QwQ-Distrill is a Chinese fine-tuning dataset constructed with reference to the SmolTalk-Chinese dataset. It aims to provide high-quality synthetic reasoning data support for training large language models (LLMs). The dataset consists entirely of synthetic data, comprising over 700,000 entries. It is specifically designed to enhance the performance of Chinese LLMs across various tasks… See the full description on the dataset page: https://huggingface.co/datasets/ChinaunicomSoftware/smoltalk-chinese-QwQ-Distrill.smollm-corpus-instruct-2M-cosmopedia-v2-gpt2-v2-streaming
A corpus of high quality fine tuning data meant for fine tuning various HelixLM models
Dataset Composition:
A subset sampled from randomly selected shards from https://huggingface.co/datasets/HuggingFaceTB/smollm-corpus cosmopedia-v2 split ...
Added: A preprocessed column formattedconversation concatenating the prompt, text and special tokens for instruct fine tuning.
Added: A token count column tokencount based on GPT2 tokenizer + the fine tuning special tokens.… See the full description on the dataset page: https://huggingface.co/datasets/david-thrower/smollm-corpus-instruct-2M-cosmopedia-v2-gpt2-v2-streaming.smol-koreantalkSmolLM2의 인스트럭션 훈련 데이터 HuggingFaceTB/smol-smoltalk를 한국어로 번역했어요.
smoleval
SmolEval
Pick the right base model before you fine-tune
Which small base model is worth your training time? There are dozens under
2B, and fine-tuning the wrong one costs hours. This scores one in a few
minutes, on what base models actually do: continue text.
90 prompts, 3-run average
coherence
relevance
diversity
SmolLM2-135M
███░░░░░░░ 34%
█░░░░░░░░░ 7%
█░░░░░░░░░ 8%
SmolLM2-360M
██░░░░░░░░ 21%
░░░░░░░░░░ 0%
█░░░░░░░░░ 14%
SmolLM2-1.7B
████░░░░░░… See the full description on the dataset page: https://huggingface.co/datasets/sifat-febo/smoleval.smolmo-proofs-cot-sft
SmolMo Proofs CoT SFT
A supervised fine-tuning dataset for teaching small open language models to write and verify rigorous mathematical proofs with chain-of-thought reasoning. It pairs proof-generation data with verification data, so a model trained on it can both solve olympiad-style problems and grade candidate solutions — supporting solve-then-verify inference workflows.
~302K examples · ~6.1B tokens · ~124K unique problems. Every example is a chat conversation [system, user… See the full description on the dataset page: https://huggingface.co/datasets/chankhavu/smolmo-proofs-cot-sft.smolgpt-markdown-stories
SmolGPT-Fables Stories
A deterministic, text-only corpus of 96,000 original English
Markdown stories built for SmolGPT-Fables. Every row is one complete supervised
story example with an exact prompt / completion boundary, a requested scene
count from one to six, and plain-language conditioning fields.
No model, API, browser, or network service was used to create this dataset.
Dataset summary
96,000 stories across 96,000 isolated story families
25 genres and all… See the full description on the dataset page: https://huggingface.co/datasets/neonforestmist/smolgpt-markdown-stories.smol-smoltalk-pt-v1
smol-smoltalk-pt
Dataset Summary
smol-smoltalk-pt is an experimental conversational SFT dataset built from the AMALIA-LLM-0626-SFT-Dataset.
The original objective was to rapidly obtain a dataset with characteristics similar to HuggingFaceTB/smol-smoltalk, while using only conversations already present in the AMALIA dataset.
The selection process attempted to approximate:
approximately 80% Portuguese and 20% English conversations;
the distribution of conversation… See the full description on the dataset page: https://huggingface.co/datasets/inaciose/smol-smoltalk-pt-v1.smoltalk-creative-writing-enPurified-openai-messages
📖 SmolTalk-Creative-Writing-enPurified-openai-messages
SmolTalk-Creative-Writing-enPurified is a highly curated, "prose-first" subset of the original collinear-ai/smoltalk-creative-writing dataset.
The enPurified collection is built on a specific philosophy: Specialization. While the ecosystem has plenty of datasets for coding (StackOverflow, StarCoder) and mathematics (GSM8K), high-quality, fluent English prose often gets diluted when mixed with syntax-heavy code or rigid math… See the full description on the dataset page: https://huggingface.co/datasets/enPurified/smoltalk-creative-writing-enPurified-openai-messages.smoltalk-gpt2-sft
SmolTalk-GPT2-SFT
A fully tokenized version of the entire HuggingFaceTB/smoltalk
(all config) dataset — converted from raw multi-turn conversations into ready-to-train
(token_id, loss_mask) pairs using the GPT-2 BPE tokenizer (vocab=50257). Drop
it straight into a PyTorch trainer for supervised fine-tuning (SFT) of any
GPT-2-vocab language model — no template parsing, no role tagging, no chat-template
gymnastics required.
Overview
Source: HuggingFaceTB/smoltalk (all… See the full description on the dataset page: https://huggingface.co/datasets/juliannunezb/smoltalk-gpt2-sft.smolified-password-strength-checker
🤏 smolified-password-strength-checker
Intelligence, Distilled.
This is a synthetic training corpus generated by the Smolify Foundry.
It was used to train the corresponding model AnirbanRay20/smolified-password-strength-checker.
📦 Asset Details
Origin: Smolify Foundry (Job ID: cf77efc0)
Records: 3000
Type: Synthetic Instruction Tuning Data
⚖️ License & Ownership
This dataset is a sovereign asset owned by AnirbanRay20.
Generated via Smolify.ai.
kakugo-smo
Kakugo Samoan dataset
[Paper] [Code] [Model]
A synthetically generated conversation dataset for training in Samoan.
This dataset contains synthetic conversational data and translated instructions designed to train Small Language Models (SLMs) for Samoan. It was generated using the Kakugo pipeline, a method for distilling high-quality capabilities from a large teacher model into low-resource language models. The teacher model used to generate this dataset was… See the full description on the dataset page: https://huggingface.co/datasets/ptrdvn/kakugo-smo.smoltalk-smol-magpie-ultra-no-refusals
SmolTalk Smol-Magpie-Ultra No Refusals
A Minos-cleaned version of HuggingFaceTB/smoltalk / smol-magpie-ultra for use as a neutral helpfulness SFT anchor.
Rows are removed when NousResearch/Minos-v1 classifies the conversation as a refusal. The original train/test split structure is preserved.
Cleaning version: minos-only-v1-2026-06-23
Counts
Split
Input rows
Kept rows
Dropped rows
train
409,537
408,447
1,090
test
21,555
21,488
67
Overall removal… See the full description on the dataset page: https://huggingface.co/datasets/nchapman/smoltalk-smol-magpie-ultra-no-refusals.