datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
VideoChat-Flash-Training-Data
🦜 VideoChat-Flash-Training-Data
This repos contains all annotaions and most videos for training VideoChat-Flash.
📕 How to use the LongVid data?
For video_dir like longvid_subset/coin_grounding_10k_zip, you need to concat this dir to a zip file as follows:
cat ego4dhcap_eventunderstanding_2k_zip/* > ego4dhcap_eventunderstanding_2k.zip
✏️ Citation
@article{li2024videochatflash,
title={VideoChat-Flash: Hierarchical Compression for Long-Context… See the full description on the dataset page: https://huggingface.co/datasets/OpenGVLab/VideoChat-Flash-Training-Data.GLM-5.3-Flash-TR3-partsbin-v1
GLM-5.3-Flash TR3 parts bin v1 — K6 + K8 payload stores under one transform seed
This dataset is the parts bin for the GLM-5.3-Flash TR3 quantization
campaign (2026-08-27/28): the complete per-choice payload stores of the two
published uniform quants, plus the preparation artifacts and provenance
receipts that produced them.
malaiwah/GLM-5.3-Flash-TR3-6bpw (uniform K6)
malaiwah/GLM-5.3-Flash-TR3-8bpw (uniform K8)
What a parts bin is
TR3 (trellis) encoding is… See the full description on the dataset page: https://huggingface.co/datasets/malaiwah/GLM-5.3-Flash-TR3-partsbin-v1.FlashRAG_datasets
⚡FlashRAG: A Python Toolkit for Efficient RAG Research
FlashRAG is a Python toolkit for the reproduction and development of Retrieval Augmented Generation (RAG) research. Our toolkit includes 36 pre-processed benchmark RAG datasets and 16 state-of-the-art RAG algorithms.
With FlashRAG and provided resources, you can effortlessly reproduce existing SOTA works in the RAG domain or implement your custom RAG processes and components.
For more information, please view our GitHub repo… See the full description on the dataset page: https://huggingface.co/datasets/RUC-NLPIR/FlashRAG_datasets.flashinfer-trace
FlashInfer Trace
We provide an official dataset called FlashInfer Trace with kernels and workloads in real-world AI system deployment environments. FlashInfer-Bench can use this dataset to measure and compare the performance of kernels. It follows the FlashInfer Trace Schema.
It is organized as follows:
flashinfer_trace/ # Here
├── definitions/
└── workloads/
flashinfer-trace/ # On Hugging Face
├── solutions/
└── traces/
Example solutions and traces directories, featuring… See the full description on the dataset page: https://huggingface.co/datasets/flashinfer-ai/flashinfer-trace.medical_meadow_medical_flashcards
Dataset Card for Medical Flashcards
Dataset Summary
Medicine as a whole encompasses a wide range of subjects that medical students and graduates must master
in order to practice effectively. This includes a deep understanding of basic medical sciences, clinical knowledge,
and clinical skills. The Anki Medical Curriculum flashcards are created and updated by medical students and cover the
entirety of this curriculum, addressing subjects such as anatomy, physiology… See the full description on the dataset page: https://huggingface.co/datasets/medalpaca/medical_meadow_medical_flashcards.Step-3.5-Flash-SFT
Step-3.5-Flash-SFT
Step-3.5-Flash-SFT is a general-domain supervised fine-tuning release for chat models.
This repository keeps the full training interface in one place:
json/: canonical raw training data
tokenizers/: tokenizer snapshots for Step-3.5-Flash and Qwen3, released to preserve chat-template alignment
compiled/: tokenizer-specific compiled shards for StepTronOSS training
Data Format
Each raw shard is a JSON file whose top level is a list of examples.… See the full description on the dataset page: https://huggingface.co/datasets/stepfun-ai/Step-3.5-Flash-SFT.transformers_flash_attn_cilaions_got_talent_enhanced_flash_annotations_and_long_captionsGLM-5.3-Flash-BF16-Teacher-Logits
GLM-5.3-Flash BF16 teacher logits
This dataset contains full-vocabulary float32 teacher logits from the immutable
zai-org/GLM-5.3-Flash-BF16 revision a6c167b62691b2bac901344b65cb651a70f53e43.
It keeps the sealed final KLD panel qualification-only and publishes the
separate non-final calibration panel under role-specific paths.
Qualification-only final windows: 25
Qualification-only final prediction positions: 51175
Vocabulary size: 154880
Teacher receipt:… See the full description on the dataset page: https://huggingface.co/datasets/brandonmusic/GLM-5.3-Flash-BF16-Teacher-Logits.Flash100KMulti-SWE-bench-flash
👋 Overview
This repository contains the Multi-SWE-bench dataset, introduced in Multi-SWE-bench: A Multilingual Benchmark for Issue Resolving, to address the lack of multilingual benchmarks for evaluating LLMs in real-world code issue resolution.
Unlike existing Python-centric benchmarks (e.g., SWE-bench), this framework spans 7 languages (Java, TypeScript, JavaScript, Go, Rust, C, and C++) with 1,632 high-quality instances,
curated from 2,456 candidates by 68 expert annotators… See the full description on the dataset page: https://huggingface.co/datasets/ByteDance-Seed/Multi-SWE-bench-flash.A10_benchmark_flash_attentionlaions_got_talent_enhanced_just_flash_annotationsGLM-5.3-Flash-fidelity-suite-v1
GLM-5.3-Flash Fidelity Suite v1
Historical distribution-fidelity evidence for GLM-5.3-Flash (released 2026-08-26):
BF16-reference and FP8-as-served hidden-state captures, a shared LM head, and
receipts from the declared capture/replay path. Compatible candidate captures
can be compared on matching published positions without holding the 643 GB
reference; this is not a universal native-serving or task-quality score. Protocol: the Qwen3.8-27B fidelity-suite v5 methodology… See the full description on the dataset page: https://huggingface.co/datasets/malaiwah/GLM-5.3-Flash-fidelity-suite-v1.ioi-eval-openrouter_google_gemini-2_0-flash-thinking-exp-prompt-mem-limitosworld-glm-5.3-flash-trajThese are the trajectory results from our GLM-5.3-Flash evaluation on OSWorld.
For detailed evaluation results, configuration, and additional information, please refer to the following GitHub issue:
https://github.com/xlang-ai/OSWorld/issues/591
flashvsr-repro-outputs-v2-part1
FlashVSR 复现实验输出 — part1
FlashVSR 复现及 KV cache 驱逐策略消融实验的逐帧推理输出，以 FFV1 无损编码归档。
本 repo 是全部结果的第 1/2 部分。
内容
reds_bscv_dbl_clean_full_sliding
reds_bscv_dbl_clean_full_sliding_kv10
reds_bscv_dbl_clean_full_sliding_kv6
reds_bscv_dbl_full_gate
reds_bscv_dbl_full_gate_kv10
reds_bscv_dbl_full_gate_kv6
reds_bscv_dbl_full_gate_lfres
reds_bscv_dbl_full_gate_lfres_frame
reds_bscv_dbl_full_gate_lfres_frame_kv10
reds_bscv_dbl_full_gate_lfres_frame_kv6… See the full description on the dataset page: https://huggingface.co/datasets/victorzhu30/flashvsr-repro-outputs-v2-part1.gemini-flash-2.0-speech
🎙️ Gemini Flash 2.0 Speech Dataset
This is a high quality synthetic speech dataset generated by Gemini Flash 2.0 via the Multimodal Live API. It contains speech from 2 speakers - Puck (Male) and Kore (Female) in English.
🏅 #1 Trending Audio Dataset in Feb 2025
🏅 Used in training of Kokoro TTS and LLaSA 1B
〽️ Stats
Total number of audio files: 47,256*2 = 94512Total duration: 1023527.20seconds (284.31 hours)
Average duration: 10.83 seconds
Shortest file: 0.6… See the full description on the dataset page: https://huggingface.co/datasets/shb777/gemini-flash-2.0-speech.tb2-k5-a21h-flashtb2-k5-a21l-flashStep-3.5-Flash-SFT-code
Step-3.5-Flash-SFT-code
Code SFT dataset extracted from stepfun-ai/Step-3.5-Flash-SFT, converted to extended OpenAI SFT format, with multi-dimensional quality labels and thinking-mode classification.
602,595 conversations | 40.30 GB | 80 files
Dataset Summary
Group
Records
Size
Avg Rounds
Description
single_turn/slow
445,173
20.75 GB
1.0
Single-turn with chain-of-thought reasoning
single_turn/fast
9,746
0.12 GB
1.0
Single-turn without reasoning… See the full description on the dataset page: https://huggingface.co/datasets/liuxsh9/Step-3.5-Flash-SFT-code.tb21-dsv4-flash-0731-dsh
Terminal-Bench 2.1 trajectories: DeepSeek-V4-Flash-0731 + dsh sdk-minimal
Every trial of this one line, in one place: the 89-task main run, both re-run passes, and
the scoring scripts. The trajectories are raw and unedited — each step's reasoning, each
tool call, and the verifier's own stdout.
This is a re-packaging, not a new measurement. The same files were published before,
split across two releases, which made the line look incomplete in both: the first release
carried the… See the full description on the dataset page: https://huggingface.co/datasets/openguardrails/tb21-dsv4-flash-0731-dsh.FlashSR_weightsflashvsr-repro-outputs-part3
FlashVSR 复现实验输出 — part3
FlashVSR 复现及 KV cache 驱逐策略消融实验的逐帧推理输出，以 FFV1 无损编码归档。
本 repo 是全部结果的第 3/3 部分。
内容
reds_val30_gate
reds_val30_gate_L0-15
reds_val30_gate_L15-30
reds_val30_gate_lam1_fifo
reds_val30_gate_lam5_fifo
reds_val30_gate_lfres
reds_val30_gate_lfres_raw
reds_val30_gate_lfres_z
reds_val30_gate_sink
reds_val30_gate_tjump
reds_val30_h2o
reds_val30_headwise
reds_val30_headwise_kv10
reds_val30_ofr_gate
reds_val30_ofr_gate_L0-15
reds_val30_ofr_gate_L15-30… See the full description on the dataset page: https://huggingface.co/datasets/victorzhu30/flashvsr-repro-outputs-part3.Taur_CoT_Analysis_Project___google__gemini-1.5-flash-001GLM-5.3-Flash-calibration-activations-v1
GLM-5.3-Flash calibration activations v1 (BF16, natural routing)
Per-layer block-input activations of zai-org/GLM-5.3-Flash-BF16 @ b1967181 over 92x2048
tokens of the exllamav3 standard_cal_data corpus (pinned): per context, layer_NNN.attn_in
and layer_NNN.mlp_in (bf16, post-norm linear inputs; mlp_in is the router + expert gate/up
input) and layer_NNN.router_logits (fp32, natural top-8 routing ground truth).
Per-expert Hessians E[xx^T], routing statistics and down-proj inputs… See the full description on the dataset page: https://huggingface.co/datasets/malaiwah/GLM-5.3-Flash-calibration-activations-v1.mlsys26-contest
MLSys 2026 FlashInfer-Bench Challenge Dataset
This repository contains the FlashInfer-Bench dataset for the MLSys 2026 Kenrel Generation Challenge.
This dataset targets to be used in the FlashInfer-Bench benchmark system.
It follows the FlashInfer Trace Schema. To use the dataset in the competition, please refer to our starter kit.
Download
Use this command to download the dataset:
git lfs install
git clone https://huggingface.co/datasets/flashinfer-ai/mlsys26-contest… See the full description on the dataset page: https://huggingface.co/datasets/flashinfer-ai/mlsys26-contest.flash-flood-benchmark-data
TORRENT — CONUS Flash-Flood Benchmark (L1–L3), agent-friendly
Traceable, Observation-constrained, Rapid-Response, Episode–gauge–watershed
Network of Testbeds: reproducible flash-flood testbeds for hydrological-response
analysis and model intercomparison. This mirror carries the paper-matched
v1.0 release (companion paper: TORRENT, Earth System Science Data).
Archive of record (v1.0): https://doi.org/10.5281/zenodo.22118051
(concept DOI, always the latest version:… See the full description on the dataset page: https://huggingface.co/datasets/skyan1002/flash-flood-benchmark-data.tb2-k5-cd1-cortex-dsh-flashglm53-flash-harvest
GLM-5.3-Flash On-Policy Harvest
86,006 responses / 246,034,910 generated tokens written by
zai-org/GLM-5.3-Flash from its reference FP8 weights,
across four harvest rounds, 15 registers and both serving modes (22,016 rows carry the
model's inline <think>…</think> chain). It is on-policy text: the corpus records what the target model
actually generates, which is what a speculative-decoding drafter (EAGLE-3 / DFlash / DSpark family) has to
learn to predict. Everything here is MIT.… See the full description on the dataset page: https://huggingface.co/datasets/Zek-Takai/glm53-flash-harvest.