New this week: can you run them?
Open models released on Hugging Face in the last 7 days that are getting real attention (likes and trending), checked against 65 GPUs and Macs. The list refreshes every hour, so new releases usually show up here within an hour or two of picking up traction. Last checked 2026-11-17 23:27 UTC.
| Model | Released | Params | 4-bit · 8K | VRAM | Cheapest GPU | Smallest Mac | Likes |
|---|---|---|---|---|---|---|---|
| SenseNova Loop 9B sensenova | Nov 17, 2026 | 9.4B | 6.4 GB | runs on 8 GB+ VRAM | Arc B580 | Mac M4 16GB | 175 |
| KORMo Web 11B KORMo-VL | Nov 14, 2026 | 11B | 8.3 GB | runs on 12 GB+ VRAM | Arc B580 | Mac M4 16GB | 73 |
| SingVL S1 4B inclusionAI | Nov 11, 2026 | 4.5B | 3.6 GB | runs on 8 GB+ VRAM | Arc B580 | Mac M4 16GB | 5 |
Earlier this month
| Model | Released | Params | 4-bit · 8K | Cheapest GPU |
|---|---|---|---|---|
| Nex N2.6 nex-agi | Nov 10, 2026 | 321B | 188 GB | none in VRAM |
| MiniMax H3 Image PE MiniMaxAI | Nov 8, 2026 | 27B | 16.8 GB | RX 7900 XT |
| Arete 27B NousResearch | Nov 6, 2026 | 27B | 17.4 GB | RX 7900 XTX |
| Inkling 2 Small thinkingmachines | Nov 4, 2026 | 594B | 346 GB | none in VRAM |
| LongCat 2.5 meituan-longcat | Nov 4, 2026 | 1.78T | 1014 GB | none in VRAM |
| Kimi K3.1 Pro moonshotai | Nov 2, 2026 | 2.78T | 1584 GB | none in VRAM |
| Mage VL Cowork microsoft | Oct 30, 2026 | 4.7B | 4.3 GB | Arc B580 |
| Raon 1 Nano KRAFTON | Oct 29, 2026 | 3.5B | 3.0 GB | Arc B580 |
| SenseNova Loop 9B Preview sensenova | Oct 28, 2026 | 9.7B | 6.4 GB | Arc B580 |
| Ling 3.1 Flash 1019 inclusionAI | Oct 27, 2026 | 572B | 335 GB | none in VRAM |
| LFM3 1.7B Base LiquidAI | Oct 26, 2026 | 1.7B | 1.7 GB | Arc B580 |
| Unlimited OCR 2 baidu | Oct 26, 2026 | 3.3B | 3.0 GB | Arc B580 |
| d1 2B LiquidAI | Oct 24, 2026 | 1.9B | 1.8 GB | Arc B580 |
| LFM3 1.7B LiquidAI | Oct 23, 2026 | 1.7B | 1.8 GB | Arc B580 |
| WeVisOne VL 4B tencent | Oct 22, 2026 | 4.8B | 4.3 GB | Arc B580 |
Everything new this month, plus new API models → · RSS
How a model gets listed
Every hour we check Hugging Face's trending lists and the newest uploads from the AI labs we track. A release from the last 14 days is listed when it's from a tracked lab or has at least 20 likes or a trending score of 15. Finetunes from new accounts need 60 likes. Quantized re-uploads, merges and "uncensored" variants are skipped. Each listed model gets a model page, a "Can I run it?" page, GPU-by-GPU pages, a sitemap entry and an IndexNow ping. Numbers come from the model's own config and real GGUF sizes once they're published. Methodology.