datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Qwen3.8-27B-GGUF-metrics
Qwen3.8-27B GGUF, everything behind the numbers
This is the working record for
AtomicChat/Qwen3.8-27B-GGUF.
Every figure in that model card came from a file in here, including the ones
about other publishers' builds.
The point of publishing it is simple. A quantization comparison is only worth
reading if someone else can run it, and that needs three things nobody usually
ships: the exact reference the numbers were measured against, the exact text
they were measured on, and the… See the full description on the dataset page: https://huggingface.co/datasets/AtomicChat/Qwen3.8-27B-GGUF-metrics.gguf-models
GGUF Models Collection - 多格式版本
這個倉庫包含多種量化格式的 GGUF 模型檔案。
格式說明
格式
描述
品質
檔案大小
推薦用途
FP16
16位浮點
最高
最大
高精度推理、微調基準
Q8_0
8位量化
高
中等
高品質推理、伺服器部署
Q4_K_M
4位混合量化
良好
最小
本地部署、快速推理
轉換摘要
📊 成功轉換: 3/4 個模型
📈 成功率: 75.0%
🔧 支援格式: FP16, Q8_0, Q4_K_M
🕒 更新時間: 2025-08-29 04:34:43
模型列表
模型名
FP16
Q8_0
Q4_K_M
狀態
deepseek-1.3b-sql-final-t4x2
2569.5MB
N/A
N/A
✅ 成功
codegemma-2b-sql-coder-finetuned
4786.0MB
N/A
N/A
✅ 成功… See the full description on the dataset page: https://huggingface.co/datasets/Paul720810/gguf-models.Muse-Glimmer-30B-GGUF-metrics
Muse Glimmer 30B GGUF — raw metrics
Every log behind the numbers in
AtomicChat/Muse-Glimmer-30B-GGUF.
Published unfiltered, so any figure in the model card can be checked or disputed.
Layout
Path
Contents
kld/
llama-perplexity --kl-divergence output, per build and per corpus
bench/
llama-bench -o json
speculative/
llama-server logs with and without the drafter
layouts/
per-tensor type map of every GGUF
conversion/
convert_hf_to_gguf.py logs… See the full description on the dataset page: https://huggingface.co/datasets/AtomicChat/Muse-Glimmer-30B-GGUF-metrics.Ornith-1.5-35B-A3B-GGUF-metricswikipedia-viewerwikipedia dataset, now with viewer enabled! :D
Dataset Card for Wikipedia
Dataset Summary
Wikipedia dataset containing cleaned articles of all languages.
The datasets are built from the Wikipedia dump
(https://dumps.wikimedia.org/) with one split per language. Each example
contains the content of one full Wikipedia article with cleaning to strip
markdown and unwanted sections (references, etc.).
The articles are parsed using the mwparserfromhell tool, which can be… See the full description on the dataset page: https://huggingface.co/datasets/GGUFGuy/wikipedia-viewer.iter267-f16-gguf-splitLing-3.0-flash-GGUF-metrics
Ling-3.0-flash — quantization metrics
Everything measured while building the GGUF line for inclusionAI/Ling-3.0-flash: raw logs, per-rung numbers and the importance matrix statistics. Published so the quant table can be checked rather than trusted.
Quants live in AtomicChat/Ling-3.0-flash-GGUF.
Layout
metrics/
grid-table.json per rung: size, bpw, mean/99% KLD, top-1 agreement
kld-results.json raw parser output of every KL divergence run… See the full description on the dataset page: https://huggingface.co/datasets/AtomicChat/Ling-3.0-flash-GGUF-metrics.Ornith-1.5-9B-GGUF-metricsQwen3.8-Flash-Next-GGUF-metricsQwQ-32B-abliterated-131k-GGUF-Yarn-Imatrix
QwQ-32B-Abliterated-131k-GGUF-Yarn-Imatrix
High-Fidelity Semantic Simulation & Orchestration AI Model
Will this pass the random stupid benchmarks that exist today? I don't know, nor care. I don't need my local AI model to know some random city capital of a foreign country. I need a local AI model that can simulate with high semantic fidelity. Why? Because your AI may be able to spit random facts. I want an AI that knows when to Google facts. I want an AI that tracks hundreds of… See the full description on the dataset page: https://huggingface.co/datasets/magiccodingman/QwQ-32B-abliterated-131k-GGUF-Yarn-Imatrix.qwen36-27b-gguf-bfcl-v4-quantization-pilot-corrected-v3
Qwen3.6-27B GGUF quantization on a bounded BFCL V4 pilot
Q4_K_M matched Q8_0 on both tested categories: each scored 94 of 100 selected cases correct. Q5_K_M also scored 94/100; Q3_K_M scored 92/100.
Read the results page · Inspect all 400 scored rows
This is a post-result-corrected exploratory analysis of two selected non-live BFCL V4 categories, not a full leaderboard result.
Inspect the scored rows without cloning
The Hub Dataset Viewer does not render this… See the full description on the dataset page: https://huggingface.co/datasets/CyberNative-AI/qwen36-27b-gguf-bfcl-v4-quantization-pilot-corrected-v3.gguf-jlens-artifacts
gguf-jlens artifacts
Heavy artifacts for the gguf-jlens project (Jacobian lens
on GGUF-quantized Qwen3.5-4B) — kept out of git and mirrored here.
lenses/ — fitted Jacobian lenses (*.pt) and resume checkpoints (*.ckpt.pt):
q8_0, q4_k_m, q3_k_m, q2_k, bf16-control (+ -s2 disjoint-sample replicas).
models/ — GGUFs quantized locally from the unsloth BF16 file (Q3_K_M, Q2_K).
out/ — per-item rank dumps (ranks/, ranks5/, ranks6/), slice pages, fit logs.
Restore into a clone with… See the full description on the dataset page: https://huggingface.co/datasets/mozilla-ai/gguf-jlens-artifacts.fish-audio-s2-gguf-models
Fish Audio S2 models
This Kaggle dataset contains a rodrigomt/s2-pro-gguf checkpoint snapshot for gguf-s2-cpp PHRunner Fish Audio S2 inference.
Source repository: rodrigomt/s2-pro-gguf
Revision: main
Layout: s2-pro-gguf
Required backend: gguf-s2-cpp
Default model file: s2-pro-f16.gguf
Default codec file: ``
Files: 10
The runner expects this directory to be mounted as a Kaggle input dataset. Official Python/SGLang layouts are validated through config.json, .safetensors files… See the full description on the dataset page: https://huggingface.co/datasets/stokiz/fish-audio-s2-gguf-models.gguf-repro-harness
Two-run reproducibility of a GGUF quantization and evaluation harness: quality and memory reproduce, wall-clock latency does not
We ran the same pinned pipeline twice — two different operators, fresh containers, same commands — on two Apache-2.0 models (Qwen2.5-0.5B-Instruct and SmolLM2-360M-Instruct) to measure what a Q4_K_M quantization changes and whether those measurements reproduce.
What reproduced across both runs (exact, or within ±2%):
the F16 and Q4_K_M GGUF files are… See the full description on the dataset page: https://huggingface.co/datasets/CyberNative-AI/gguf-repro-harness.Wan_2.2_T2V_10steps_GGUFWan_2.2_I2V_10steps_GGUFagentic-safety-gguf
agentic-safety-gguf: Training & Evaluation Datasets
Model: guerilla7/agentic-safety-ggufPaper: (https://arxiv.org/abs/2601.00848)Total: 80,992 examples (80,851 after deduplication)
Overview
Complete training and evaluation datasets for agentic-safety-gguf, a specialized Llama 3.1 8B model for agentic AI security analysis. Supports iterative continuation training methodology (V2→V3→V4) for full reproducibility.
Dataset Files
File
Examples
Size
Purpose… See the full description on the dataset page: https://huggingface.co/datasets/guerilla7/agentic-safety-gguf.qwen3-5-gguf-tiny-fidelity-root-v1
qwen35-gguf random CPU fixture root
A root fidelity dataset in hidden form, produced by engines/tools/hf_capture.py from malaiwah/qwen3-5-gguf-tiny-random-bf16.
The cut
the final hidden state handed to lm_head -- after the text model's final norm and immediately before the head matmul -- captured as the head module's input via torch.nn.Module.register_forward_pre_hook; replay applies the head ONLY (no final norm at replay time: the capture already sits after it).… See the full description on the dataset page: https://huggingface.co/datasets/malaiwah/qwen3-5-gguf-tiny-fidelity-root-v1.moss-tts-v1-5-gguf-models
MOSS-TTS model payload
This dataset was generated by MossTtsModelsUploader for PHRunner.Kaggle.Service.MossTtsGguf.
Source
Hugging Face repository: mmmsss1234/openmoss-moss-tts-v15-t4
Revision: main
Kaggle dataset: kaggle-pool-account/moss-tts-v1-5-gguf-models
Layout: openmoss-ggml-v15-t4
Required backend: openmoss-ggml-cli
Default model file: moss-tts-1.5-q4km.gguf
Files
moss-tts-1.5-q4km.extras.gguf - sidecar - 3.81 GiB - SHA256… See the full description on the dataset page: https://huggingface.co/datasets/stokiz/moss-tts-v1-5-gguf-models.flux_gguf
ComfyUI Workflow - Flux GGUF
Workflow description
This is an all in one workflow of all my previous workflows, which includes text to image, image to image, inpainting, and controlnet, a multi LoRA loader and a Finetune/Upscale option that can be disabled. All condensed in one workflow so you do not have to switch between them.
Download workflow
Download the files in the Files & Versions tab.
moss-tts-gguf-runtime-portable
MOSS-TTS runtime payload
This private Kaggle dataset is generated by Phorcys.Tools.MossTtsGgufRuntimeUploader for PHRunner.Kaggle.Service.MossTtsGguf.
Runtime flavor: LinuxCuda
Python tag: python3.10
Wheelhouse mode: not staged
Generated UTC: 2026-09-08T11:01:08.9030626+00:00
The dataset intentionally contains runtime artifacts, not the GGUF model repository by default. Keep the model files in a separate private Kaggle dataset, for example… See the full description on the dataset page: https://huggingface.co/datasets/stokiz/moss-tts-gguf-runtime-portable.voxcpm2-gguf-models
VoxCPM2 GGUF models
This Kaggle dataset contains VoxCPM2 GGUF model files for the VoxCPM.cpp/GGUF backend.
Source repository: bluryar/VoxCPM-GGUF
Revision: main
Layout: gguf
Required backend: gguf-voxcpm-cpp
Default model file: voxcpm2-f16.gguf
Files: 3
The runner expects the full directory to be mounted as a Kaggle input dataset. HF/Nano-vLLM/vLLM-Omni layouts are discovered by config.json, model.safetensors, and audiovae.pth or audiovae.safetensors; GGUF layouts are… See the full description on the dataset page: https://huggingface.co/datasets/stokiz/voxcpm2-gguf-models.glm53-fidelity-gguf-unsloth-udq4kxl-v1
fidelity--glm53.malaiwah.quant.gguf-unsloth-udq4kxl
A quant fidelity dataset in hidden form, produced by engines/tools/hf_capture.py from unsloth/GLM-5.3-GGUF.
The cut
the final hidden state handed to lm_head -- after the text model's final norm and immediately before the head matmul -- captured as the head module's input via torch.nn.Module.register_forward_pre_hook; replay applies the head ONLY (no final norm at replay time: the capture already sits after it).… See the full description on the dataset page: https://huggingface.co/datasets/malaiwah/glm53-fidelity-gguf-unsloth-udq4kxl-v1.qfs-qwen-gguf-tiny-cpu-format-v1
qwen-gguf tiny CPU format fixture reproducibility
Complete tiny random FORMAT fixture evidence. Round-to-nearest (RTN) storage/reader exercise only; optimizer-not-run. No GPTQ/AWQ/AutoRound optimization, calibrated ModelOpt/CT/QAT quality, trained-model quality ranking, GPU parity, or native serving-kernel correctness claim.
Reconstructed weights are evaluated by the captured native forward. KL is own-head, full-vocabulary on the recorded panel, not a benchmark of training… See the full description on the dataset page: https://huggingface.co/datasets/malaiwah/qfs-qwen-gguf-tiny-cpu-format-v1.voxcpm2-gguf-runtime-portable
VoxCPM2 runtime payload
This private Kaggle dataset is generated by Phorcys.Tools.VoxCPM2RuntimeUploader for PHRunner.Kaggle.Service.VoxCPM2.
Runtime flavor: LinuxCuda
Python tag: python3.10
Generated UTC: 2026-09-09T13:31:02.3368864+00:00
The dataset intentionally contains runtime artifacts, not model weights. Keep the official openbmb/VoxCPM2 checkpoint snapshot in a separate private Kaggle dataset, for example kaggle-pool-account/voxcpm2-python-models.
Top-level runtime… See the full description on the dataset page: https://huggingface.co/datasets/stokiz/voxcpm2-gguf-runtime-portable.dolia-tts-qwen3-tts-lora-registry-gguf
A-walla-walla/dolia-tts-qwen3-tts-lora-registry-gguf
qwen3-tts.cpp gguf voice registry — the input manifest the server consumes. Maps
2 voice name(s) to their published adapter gguf, with a backlink to the source
registry it was generated from.
Source registry
A-walla-walla/dolia-tts-qwen3-tts-lora-registry
Source revision
2340956ddbb290e9fcd8d73b656220cfd45ec735
Voices
2
Generated by scripts/publish_gguf_registry.py.
voxcpm2-gguf-runtime-cpu-portable
VoxCPM2 runtime payload
This private Kaggle dataset is generated by Phorcys.Tools.VoxCPM2RuntimeUploader for PHRunner.Kaggle.Service.VoxCPM2.
Runtime flavor: LinuxCpu
Python tag: python3.10
Generated UTC: 2026-09-09T14:18:24.2697296+00:00
The dataset intentionally contains runtime artifacts, not model weights. Keep the official openbmb/VoxCPM2 checkpoint snapshot in a separate private Kaggle dataset, for example kaggle-pool-account/voxcpm2-python-models.
Top-level runtime… See the full description on the dataset page: https://huggingface.co/datasets/stokiz/voxcpm2-gguf-runtime-cpu-portable.glm52-fidelity-gguf-unsloth-udq4kxl-v1
fidelity--glm52.malaiwah.quant.gguf-unsloth-udq4kxl
A quant fidelity dataset in hidden form, produced by engines/tools/hf_capture.py from unsloth/GLM-5.2-GGUF.
The cut
the final hidden state handed to lm_head -- after the text model's final norm and immediately before the head matmul -- captured as the head module's input via torch.nn.Module.register_forward_pre_hook; replay applies the head ONLY (no final norm at replay time: the capture already sits after it).… See the full description on the dataset page: https://huggingface.co/datasets/malaiwah/glm52-fidelity-gguf-unsloth-udq4kxl-v1.details_Josephgflowers__Tinyllama-616M-Cinder-DPO-With-GGUF
Dataset Card for Evaluation run of Josephgflowers/Tinyllama-616M-Cinder-DPO-With-GGUF
Dataset automatically created during the evaluation run of model Josephgflowers/Tinyllama-616M-Cinder-DPO-With-GGUF on the Open LLM Leaderboard.
The dataset is composed of 63 configuration, each one coresponding to one of the evaluated task.
The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the… See the full description on the dataset page: https://huggingface.co/datasets/open-llm-leaderboard-old/details_Josephgflowers__Tinyllama-616M-Cinder-DPO-With-GGUF.NVIDIA-Nemotron-3.5-Lightning-30B-A3B-GGUF-metrics