Daily download tracking since 28 Jul 2026

Which open models are actually winning — measured daily, not guessed.

We snapshot downloads, likes and file inventories for the top 1,030 Hugging Face models every night at 04:00 UTC. That history powers trend charts, VRAM estimates and side-by-side comparisons you can't get from the model card.

Catalog downloads · trailing 30d
1.8 B
▲ 2.4% vs previous 30d
last 40 days
25,487
compare pages
0
real outputs
40
days tracked
#Model30dΔ 7dTrend
01 glm-4-9b-chat-IMat-GGUF legraphista · text-generation 871 K ▲ 318.3%
02 NVIDIA-Nemotron-3-Nano-4B-BF16 nvidia · text-generation 1.8 M ▲ 315.7%
03 Llama-3.3-70B-Instruct meta-llama · text-generation 816 K ▲ 184.0%
04 granite-4.1-30b ibm-granite · text-generation 486 K ▲ 174.6%
05 Juggernaut-XL-v9 RunDiffusion · text-to-image 593 K ▲ 165.6%
06 Ornith-1.5-397B-GGUF ornith-ai · text-generation 740 K ▲ 160.9%
07 Qwen3.8-4B-Distill-GGUF empero-ai · text-generation 445 K ▲ 154.3%
08 FLUX.2-klein-4B-mflux-4bit Runpod · text-to-image 58 K ▲ 148.7%

Fresh comparisons

For machines
Read-only JSON, no key, 60 req/min. Snapshots also ship as a daily CSV dump.
Sponsored · GPU cloud
Rent an A100 by the hour
Run 70B models without buying hardware. From $1.29/hr.
Data from the public Hugging Face API, snapshotted daily. VRAM figures are estimates — see /methodology.