Trailing 30-day download ranking of Hugging Face language models (text-generation and image-text-to-text). Global Top 300 and Korea Top 30, country share (US 26.8%, China 70.7%). Updated daily, as of 2026-11-19. Korea #1: POCKET-Darwin-180B-GGUF.
허깅페이스 언어모델(text-generation·image-text-to-text)의 최근 30일 다운로드 순위. 글로벌 Top 300과 한국 조직 Top 30, 국가별 점유율(미국 26.8%, 중국 70.7%). 매일 갱신, 2026-11-19 기준. 한국 1위 POCKET-Darwin-180B-GGUF.
Ranking axis: trailing 30-day downloads on Hugging Face language models, text-generation and image-text-to-text (not cumulative). Country is attributed by model family (who trained it), not the uploading organization, in the global tabs. The Korea tab counts only models published directly by the official account of their Korean developer; conversions re-uploaded under the original name are excluded. CI test fixtures and embedding/reranker repositories are excluded. Data from the Hugging Face API, refreshed daily.
| # | Country | Org | Model | 30-day downloads | All-time |
|---|---|---|---|---|---|
| 1 | US | openai-community | openai-community/gpt2 | 34,439,416 | 978,783,696 |
| 2 | CN | Qwen | Qwen/Qwen3-0.6B | 16,044,059 | 264,835,303 |
| 3 | CN | Qwen | Qwen/Qwen3-14B | 13,223,290 | 50,307,400 |
| 4 | CN | deepseek-ai | deepseek-ai/DeepSeek-V3.2 | 9,854,997 | 39,829,750 |
| 5 | CN | ISTA-DASLab | ISTA-DASLab/Qwen3.8-Flash-Next-GSQ-RCO-GGUF | 7,298,936 | 13,426,348 |
| 6 | CN | Qwen | Qwen/Qwen3.8-27B | 6,888,351 | 23,916,890 |
| 7 | CN | Qwen | Qwen/Qwen2.5-7B-Instruct | 6,522,736 | 196,090,807 |
| 8 | CN | Qwen | Qwen/Qwen3.5-9B | 6,514,100 | 77,427,167 |
| 9 | CN | zai-org | zai-org/GLM-5.3-Flash | 6,378,401 | 16,191,664 |
| 10 | CN | Qwen | Qwen/Qwen3-VL-8B-Instruct | 6,361,206 | 80,989,619 |
| 11 | CN | Qwen | Qwen/Qwen3.8-27B-FP8 | 6,285,486 | 19,940,124 |
| 12 | US | google/gemma-4-31B-it | 6,264,804 | 74,878,374 | |
| 13 | CN | Qwen | Qwen/Qwen3-8B | 6,060,361 | 139,103,127 |
| 14 | CN | ornith-ai | ornith-ai/Ornith-1.5-9B-GGUF | 5,679,963 | 16,849,231 |
| 15 | US | openai | openai/gpt-oss-20b | 5,576,861 | 110,106,793 |
| 16 | CN | Qwen | Qwen/Qwen3.5-2B | 5,503,788 | 25,285,358 |
| 17 | CN | Qwen | Qwen/Qwen2.5-VL-7B-Instruct | 5,362,120 | 111,311,421 |
| 18 | CN | huihui-ai | huihui-ai/Huihui-Qwen3.6-35B-A3B-abliterated-MTP-GGUF | 5,119,599 | 5,531,859 |
| 19 | US | google/gemma-4-26B-A4B-it | 5,022,424 | 76,999,594 | |
| 20 | CN | Qwen | Qwen/Qwen2.5-1.5B-Instruct | 4,991,225 | 229,354,488 |
| 21 | CN | dphn | dphn/dolphin-2.9.1-yi-1.5-34b | 4,836,627 | 87,439,818 |
| 22 | US | facebook/opt-125m | 4,750,587 | 224,593,512 | |
| 23 | CN | Qwen | Qwen/Qwen3.5-4B | 4,674,064 | 58,369,936 |
| 24 | CN | ISTA-DASLab | ISTA-DASLab/DeepSeek-V4.1-Flash-1017-GSQ-RCO-GGUF | 4,388,749 | 4,388,749 |
| 25 | US | google/gemma-4-E4B-it | 4,354,817 | 40,455,462 | |
| 26 | CN | huihui-ai | huihui-ai/Huihui-Qwen3.8-27B-abliterated-GGUF | 4,351,329 | 10,064,702 |
| 27 | CN | SC117 | SC117/Qwen3.8-Flash-Next-GSQ-RCO-abliterated-GGUF | 4,217,023 | 6,165,750 |
| 28 | US | farbodtavakkoli | farbodtavakkoli/OTel-2.0-Decision-26B-A4B-4bit | 4,194,624 | 4,194,799 |
| 29 | US | meta-llama | meta-llama/Llama-3.1-8B-Instruct | 4,149,614 | 194,822,170 |
| 30 | CN | nvidia | nvidia/Qwen3.6-35B-A3B-NVFP4 | 4,147,961 | 43,278,036 |
| 31 | CN | Qwen | Qwen/Qwen2.5-0.5B-Instruct | 4,098,275 | 79,157,477 |
| 32 | CN | ornith-ai | ornith-ai/Ornith-1.5-35B-A3B-GGUF | 4,083,220 | 12,830,436 |
| 33 | CN | Qwen | Qwen/Qwen3.6-35B-A3B-FP8 | 4,042,773 | 50,400,263 |
| 34 | CN | Qwen | Qwen/Qwen3.6-35B-A3B | 3,952,623 | 35,914,830 |
| 35 | CN | Qwen | Qwen/Qwen3-32B | 3,878,902 | 64,331,761 |
| 36 | CN | cdiamond | cdiamond/Qwen3.8-27B-iMatrix-NVFP4-MTP-GGUF | 3,327,159 | 10,607,984 |
| 37 | CN | Qwen | Qwen/Qwen3-VL-4B-Instruct | 3,127,953 | 33,831,207 |
| 38 | CN | Qwen | Qwen/Qwen3-4B | 3,076,428 | 95,533,672 |
| 39 | US | EleutherAI | EleutherAI/pythia-160m | 3,028,559 | 32,832,206 |
| 40 | CN | Qwen | Qwen/Qwen3.8-Flash-Next | 2,968,190 | 6,263,751 |
| 41 | CN | unsloth | unsloth/Qwen3.8-Flash-Next-GGUF | 2,902,027 | 5,626,585 |
| 42 | CN | peculiar-ragdoll | peculiar-ragdoll/Tiel-Coder-35B-A3B-GGUF-MTP | 2,832,962 | 5,734,962 |
| 43 | US | google/gemma-4-E2B-it | 2,820,414 | 24,617,359 | |
| 44 | CN | Qwen | Qwen/Qwen3-4B-Instruct-2507 | 2,773,401 | 67,494,261 |
| 45 | CN | Qwen | Qwen/Qwen3.5-0.8B | 2,772,664 | 24,281,050 |
| 46 | CN | Qwen | Qwen/Qwen3-1.7B | 2,708,669 | 69,200,481 |
| 47 | CN | lmstudio-community | lmstudio-community/Qwen3.8-27B-MLX-4bit | 2,650,520 | 12,544,824 |
| 48 | CN | unsloth | unsloth/Qwen3.6-35B-A3B-MTP-GGUF | 2,591,644 | 7,362,230 |
| 49 | US | unsloth | unsloth/gemma-4-12b-it-GGUF | 2,573,391 | 7,621,416 |
| 50 | CN | lmstudio-community | lmstudio-community/Qwen3.8-27B-MLX-8bit | 2,496,531 | 11,929,967 |
Full interactive table (Top 300, country tabs, charts): https://vidraft-global-llm-leaderboard.static.hf.space/ · Source: https://huggingface.co/spaces/VIDraft/global-llm-leaderboard