Updated 2026-11-10 · ranked by real download data
Trending text-generation
Ranked nightly from our download snapshots of the Hugging Face catalog. Every entry shows its licence and what hardware it realistically needs.
| # | Model | Params | Context | Commercial use | 30d | Min VRAM |
|---|---|---|---|---|---|---|
| 01 | Diplodocus-preview-GGUF | — | — | ✓ apache-2.0 | 663 K | — |
| 02 | LFM2.5-24B-A2B-Uncensored-GGUF | 24.0 B | 131 K | other | 639 K | from 56.9 GB |
| 03 | Underdog-Saluki-27B-1.0-MTP-Abliterated-GGUF | 27.0 B | — | ✓ apache-2.0 | 325 K | from 64.0 GB |
| 04 | Qwen3.8-27B-Defiance-Uncensored-MTP-GGUF | 27.0 B | — | ✓ apache-2.0 | 590 K | from 64.0 GB |
| 05 | gpt2-mini | 39 M | — | ✓ mit | 285 K | from 0.6 GB |
| 06 | Spark-X2.6-4B-GGUF | 4.0 B | — | ✓ apache-2.0 | 606 K | from 9.9 GB |
| 07 | Sage-A1.5-Q8_0-GGUF | — | — | ✓ apache-2.0 | 490 K | — |
| 08 | LFM2.5-24B-A2B-Turbo-Brilliance-50X-NEO-MAX-Imatrix-GGUF | 24.0 B | 131 K | other | 881 K | from 56.9 GB |
| 09 | LFM2.5-24B-A2B-GGUF | 24.0 B | 131 K | other | 430 K | from 56.9 GB |
| 10 | LFM3-1.7B-GGUF | 1.7 B | — | other | 283 K | from 4.5 GB |
| 11 | OTel-2.0-LLM-31B-IT-QLoRA | 31.3 B | — | ✓ apache-2.0 | 1.2 M | from 74.0 GB |
| 12 | NVIDIA-Nemotron-3.5-Bolt-E4B-GGUF | 4.0 B | — | other | 481 K | from 9.9 GB |
| 13 | Sage-A1.5-Q4_K_M-GGUF | — | — | ✓ apache-2.0 | 586 K | — |
| 14 | GLM-5.3-Flash-Maya-GGUF | 321.3 B | — | ✓ mit | 1.2 M | from 755.6 GB |
| 15 | OTel-2.0-Decision-26B-A4B-4bit | 25.8 B | — | ✓ apache-2.0 | 3.2 M | from 61.1 GB |
| 16 | Qwen3-Coder-30B-A3B-Instruct | 30.5 B | 262 K | ✓ apache-2.0 | 1.6 M | from 72.2 GB |
| 17 | OTel-2.0-LLM-31B-IT | 31.3 B | 262 K | ✓ apache-2.0 | 720 K | from 74.0 GB |
| 18 | GaMS3-12B-Instruct | 11.8 B | 131 K | ⚠ gemma | 226 K | from 28.2 GB |
| 19 | Step-5-Preview | 604.3 B | — | other | 7 K | from 1,420.7 GB |
| 20 | Spark-X2.5-4B-Heretic-jp-gguf | 4.0 B | — | ✓ apache-2.0 | 556 K | from 9.9 GB |
| 21 | Ornith-1.5-122B-GGUF | 122.0 B | — | ✓ mit | 871 K | from 287.2 GB |
| 22 | DeepSeek-V4.1-Flash-1017-GSQ-RCO-GGUF | 763.2 B | — | ✓ mit | 3.3 M | from 1,794.0 GB |
| 23 | humanizer | 12.0 B | — | ✓ apache-2.0 | 380 K | from 28.6 GB |
| 24 | Qwen2.5-3B | 3.1 B | 33 K | other | 203 K | from 7.8 GB |
| 25 | Beam-1.0-501B-A23B | 501.3 B | 1.0 M | ✓ apache-2.0 | 13 K | from 1,178.5 GB |
Membership and ranking refresh nightly after the snapshot run. VRAM is an estimate for the smallest quantization at 8K context — methodology.