Models
Datasets
Spaces
Buckets
Docs
Enterprise
Pricing
- Website
- Community
- Solutions
Log In
Sign Up

Collections

Discover the best community collections!

Collections trending this week

Open source Jailbreak

OBLITERATUS/Qwen3.8-27B-OBLITERATED

Text Generation • 28B • Updated Aug 24 • 533k • • 1.65k

SulphurAI/Sulphur-2-base

Text-to-Video • 9B • Updated Aug 5 • 102k • 2.22k

Kimi K3 (2.8T params, 896 experts) GGUF quants, both verified running on one 8xH100 node. Includes a sub-512 GiB tier.

qtum/Kimi-K3-GGUF

Text Generation • 2.8T • Updated Aug 17 • 807

RevGen — asset pipeline

Running on Zero

Agents

340

BiRefNet Demo

👁

340

Remove background from images with high-quality segmentation

Categoric VLM: Korean Image-Math LoRA Adapters

distill32 is the primary validated adapter; math8b_mm documents category-dependent trade-offs and is not a general upgrade.

TakTak/qwen3-vl-8b-distill32-image-math-lora

Image-Text-to-Text • Updated Aug 18 • 7
TakTak/qwen3-vl-8b-mathmm-tradeoff-research-lora

Image-Text-to-Text • Updated Aug 18 • 10

mlx-community/Qwen3.8-27B-8bit

Image-Text-to-Text • 27B • Updated Aug 14 • 27.7k • 88
FlatFootInternational/Qwen3.8-27B-mlx-6bit

Image-Text-to-Text • 27B • Updated Aug 19 • 50
Qwen/Qwen3.8-27B

Image-Text-to-Text • 28B • Updated 3 days ago • 7.51M • • 19.5k
mlx-community/Qwen3.8-27B-4bit

Image-Text-to-Text • 27B • Updated Sep 14 • 332k • 236

Sub-4-bit GGUF quants of DeepSeek-V4: Flash-0731 (seven PPL-tested tiers + DSpark draft) and Pro-0813 (1.57T, factory FP4 master).

qtum/DeepSeek-V4-Flash-0731-GGUF

Text Generation • 284B • Updated Aug 17 • 141 • 2
qtum/DeepSeek-V4-Pro-0813-GGUF

Text Generation • 1.6T • Updated Aug 21 • 1.55k • 1

Qwen3 4B to 32B in every format we ship: imatrix GGUF (seven tiers each) for llama.cpp, plus FP8, AWQ and GPTQ for vLLM.

qtum/Qwen3-4B-GGUF

Text Generation • 4B • Updated Aug 28 • 1.1k
qtum/Qwen3-8B-GGUF

Text Generation • 8B • Updated Aug 17 • 775 • 1
qtum/Qwen3-14B-GGUF

Text Generation • 15B • Updated Aug 17 • 216
qtum/Qwen3-30B-A3B-GGUF

Text Generation • 31B • Updated Aug 17 • 794 • 1

OWSM v3.1: Better and Faster Open Whisper-Style Speech Models based on E-Branchformer

Paper • 2401.16658 • Published Jan 30, 2024 • 14
Reproducing Whisper-Style Training Using an Open-Source Toolkit and Publicly Available Data

Paper • 2309.13876 • Published Sep 25, 2023 • 2
OWSM v4: Improving Open Whisper-Style Speech Models via Data Scaling and Cleaning

Paper • 2506.00338 • Published May 31, 2025 • 10

ideogram-ai/ideogram-4-fp8

Text-to-Image • 9B • Updated Jun 3 • 67.5k • • 868
aoxo/flux.1dev-abliteratedv2

Text-to-Image • 12B • Updated Mar 10 • 115 • 85

Open source Jailbreak

OBLITERATUS/Qwen3.8-27B-OBLITERATED

Text Generation • 28B • Updated Aug 24 • 533k • • 1.65k

mlx-community/Qwen3.8-27B-8bit

Image-Text-to-Text • 27B • Updated Aug 14 • 27.7k • 88
FlatFootInternational/Qwen3.8-27B-mlx-6bit

Image-Text-to-Text • 27B • Updated Aug 19 • 50
Qwen/Qwen3.8-27B

Image-Text-to-Text • 28B • Updated 3 days ago • 7.51M • • 19.5k
mlx-community/Qwen3.8-27B-4bit

Image-Text-to-Text • 27B • Updated Sep 14 • 332k • 236

SulphurAI/Sulphur-2-base

Text-to-Video • 9B • Updated Aug 5 • 102k • 2.22k

Sub-4-bit GGUF quants of DeepSeek-V4: Flash-0731 (seven PPL-tested tiers + DSpark draft) and Pro-0813 (1.57T, factory FP4 master).

qtum/DeepSeek-V4-Flash-0731-GGUF

Text Generation • 284B • Updated Aug 17 • 141 • 2
qtum/DeepSeek-V4-Pro-0813-GGUF

Text Generation • 1.6T • Updated Aug 21 • 1.55k • 1

Kimi K3 (2.8T params, 896 experts) GGUF quants, both verified running on one 8xH100 node. Includes a sub-512 GiB tier.

qtum/Kimi-K3-GGUF

Text Generation • 2.8T • Updated Aug 17 • 807

Qwen3 4B to 32B in every format we ship: imatrix GGUF (seven tiers each) for llama.cpp, plus FP8, AWQ and GPTQ for vLLM.

qtum/Qwen3-4B-GGUF

Text Generation • 4B • Updated Aug 28 • 1.1k
qtum/Qwen3-8B-GGUF

Text Generation • 8B • Updated Aug 17 • 775 • 1
qtum/Qwen3-14B-GGUF

Text Generation • 15B • Updated Aug 17 • 216
qtum/Qwen3-30B-A3B-GGUF

Text Generation • 31B • Updated Aug 17 • 794 • 1

RevGen — asset pipeline

Running on Zero

Agents

340

BiRefNet Demo

👁

340

Remove background from images with high-quality segmentation

OWSM v3.1: Better and Faster Open Whisper-Style Speech Models based on E-Branchformer

Paper • 2401.16658 • Published Jan 30, 2024 • 14
Reproducing Whisper-Style Training Using an Open-Source Toolkit and Publicly Available Data

Paper • 2309.13876 • Published Sep 25, 2023 • 2
OWSM v4: Improving Open Whisper-Style Speech Models via Data Scaling and Cleaning

Paper • 2506.00338 • Published May 31, 2025 • 10

Categoric VLM: Korean Image-Math LoRA Adapters

distill32 is the primary validated adapter; math8b_mm documents category-dependent trade-offs and is not a general upgrade.

TakTak/qwen3-vl-8b-distill32-image-math-lora

Image-Text-to-Text • Updated Aug 18 • 7
TakTak/qwen3-vl-8b-mathmm-tradeoff-research-lora

Image-Text-to-Text • Updated Aug 18 • 10

ideogram-ai/ideogram-4-fp8

Text-to-Image • 9B • Updated Jun 3 • 67.5k • • 868
aoxo/flux.1dev-abliteratedv2

Text-to-Image • 12B • Updated Mar 10 • 115 • 85

Previous
1
...
2,252
2,253
2,254
2,255
2,256
...
24,710
Next

Company

TOS Privacy About Careers

Website

Models Datasets Spaces Pricing Docs