Models · daily board

What's moving on the Hub

5,245 models with real traffic were created in the last seven days. What is trending, what is being pulled faster than it was, which base models the field is building on, what arrived this week and what everyone already uses — each with its size, licence and lineage, read nightly from every model that has traffic.

203,680 models with traffic · read 2026-09-18 · 8 days of download and like history since 2026-09-10

New models per day

2026-07-2037,248 over 60 · peak 1,060 on 2026-09-142026-09-17

Models that reached the traffic tier, by the day they were created. A spike is a release; the plateau is the field's daily rate.

#	Model	Trend score	Downloads 30d	Likes	Params	Commercial
1	Edge0/Edge0-35B-A3B-preview text-generation	2,441	37k	3,267	35B	yes
2	deepseek-ai/DeepSeek-V4.1-Flash image-text-to-text	1,666	391k	2,943	763B	yes
3	Qwen/Qwen3.8-27B image-text-to-text	609	7.5M	15,482	28B	yes
4	m-a-p/YuE2-3B text-to-audio	595	12k	685	3.6B	no
5	openbmb/MiniCPM5-2B text-generation	454	330k	1,527	2.5B	yes
6	ISTA-DASLab/Qwen3.8-27B-GSQ-RCO-GGUF image-text-to-text	432	1.0M	1,232	28B base model	yes
7	Lightricks/LTX-2.5 image-to-video	423	1.6M	4,173	—	unknown
8	TokenRhythm/NeoHorse-1-4B text-generation	419	20k	2,199	4.2B	yes
9	DavidAU/Qwen3.8-27B-TURBO-Fable-Cold-Fusion-735-882-Heretic-Uncensored-NEO-CODER-MAX-MTP-GGUF image-text-to-text	352	1.1M	826	28B base model	yes
10	ukisai/Swift-Qwen3.8-27b image-text-to-text	347	3,221	358	28B	unknown
11	unsloth/Qwen3.8-27B-GGUF	326	8.2M	4,243	28B base model	yes
12	sentence-transformers/all-MiniLM-L6-v2 sentence-similarity	260	256M	6,040	23M	yes
13	tencent/AuK text-to-speech	230	3,024	284	—	yes
14	Qwen/Qwen3.8-Flash-Next image-text-to-text	222	706k	5,343	180B	unknown
15	dealignai/DeepSeek-V4.1-Flash-UNCENSORED-FP8 image-text-to-text	221	32k	255	763B	yes
16	MiniMaxAI/MiniMax-H3 image-text-to-video	218	4.6M	5,404	33B	unknown
17	meta-llama/Llama-3.1-8B-Instruct text-generation	211	5.9M	7,681	8.0B	conditional
18	openai-community/gpt2 text-generation	205	16M	4,131	137M	yes
19	Agnes-AI/Agnes-3.0-Flash image-text-to-text	205	1,223	209	33B	yes
20	harshatheg/Qwen-2.5-1B-RLCD text-generation	202	0	206	1.5B base model	yes

Trend score is Hugging Face's own, read live. Downloads are the Hub's rolling thirty-day count. The licence column is the resolved verdict, not the card's string.

How the corpus is made

unstated68k+116%

text-generation66k+114%

text-to-image17k+24.4%

image-text-to-text14k+20.5%

sentence-similarity7,101-9.3%

automatic-speech-recognition3,959+53.0%

text-classification3,721+379%

text-to-speech3,021+62.6%

feature-extraction2,542+179%

token-classification2,208+194%

image-classification1,878+212%

robotics1,486+523%

translation1,183+45.5%

image-to-image1,162-12.0%

text-to-video1,005-5.2%

fill-mask928+72.7%

0share of 194,751 models · pill: last 30 days vs the 30 before68k

Lifetime share on the bar; the pill is the last thirty days of arrivals against the thirty before, so a category growing faster than the corpus is where the new models are going. Size classes come from parameter counts; "unstated" is a model whose card and weights give none.

Which of these you can run under which runtime, and under what terms, is worked through in Which models can you actually run? and What you're allowed to ship.