- Home
- /Open models
>>
Run any open model on Lium
Today's trending open models, with the VRAM each needs and the cheapest GPU setup that runs it right now. Pick one, rent the GPUs in seconds, pay per second.
| Model | Parameters | VRAM needed | Cheapest setup | Per hour | |
|---|---|---|---|---|---|
| Qwen3.8-27B Alibaba Qwen · Vision and text | 27.8B | 67 GB | A100 80GB PCIe | from $0.81 | Rent |
| Kimi-K3.1-Pro Moonshot AI · Vision and text | 2.8T | 1874 GB | 8x B300 SXM6 | from $32.00 | Rent |
| DeepSeek-V4.1-Flash-1017 DeepSeek · Vision and text | 763.2B | 613 GB | 8x RTX PRO 6000 Workstation | from $15.92 | Rent |
| Meta-Llama-3-8B-Instruct Meta · Text generation | 8B | 20 GB | RTX 3090 | from $0.16 | Rent |
| Qwen3.8-Flash-Next Alibaba Qwen · Vision and text | 180B | 432 GB | 8x RTX PRO 6000 Workstation | from $15.92 | Rent |
| LFM3-1.7B Liquid AI · Text generation | 1.7B | 5 GB | GTX 1660 Ti | from $0.10 | Rent |
| LFM3-Omni-2B Liquid AI · Multimodal | 2.2B | 6 GB | GTX 1660 Ti | from $0.10 | Rent |
| GLM-5.3-Flash Z.ai · Vision and text | 321.3B | 394 GB | 8x RTX PRO 6000 Workstation | from $15.92 | Rent |
| MiniMax-H3-Image-PE MiniMax · Text generation | 27.4B | 22 GB | RTX 3090 | from $0.16 | Rent |
| flux-3-image-klein Black Forest Labs · Image editing | 5.5B | 25 GB | 2x RTX 5060 Ti | from $0.30 | Rent |
| Qwen3.8-27B-DSpine-b16 Tencent · Text generation | 2B | 5 GB | GTX 1660 Ti | from $0.10 | Rent |
| FLUX.1-dev Black Forest Labs · Image generation | 11.9B | 41 GB | 2x RTX 3090 | from $0.36 | Rent |
| FLUX.1-schnell Black Forest Labs · Image generation | 11.9B | 41 GB | 2x RTX 3090 | from $0.36 | Rent |
| Llama-3.1-8B-Instruct Meta · Text generation | 8B | 20 GB | RTX 3090 | from $0.16 | Rent |
| GLM-5.3 Z.ai · Text generation | 753.3B | 907 GB | 8x H200 | from $28.00 | Rent |
| Qwen-Image-2.1 Alibaba Qwen · Image generation | 7.1B | 40 GB | 2x RTX 3090 | from $0.36 | Rent |
| d1-2B Liquid AI · Vision and text | 1.9B | 5 GB | GTX 1660 Ti | from $0.10 | Rent |
| gemma-4-31B-it Google · Vision and text | 31.3B | 75 GB | A100 80GB PCIe | from $0.81 | Rent |
| Kimi-K3 Moonshot AI · Vision and text | 2.8T | 1874 GB | 8x B300 SXM6 | from $32.00 | Rent |
| Qwen3.5-9B Alibaba Qwen · Vision and text | 9.7B | 24 GB | RTX 3090 | from $0.16 | Rent |
| MiniMax-M3.1-Flash MiniMax · Vision and text | 220.5B | 301 GB | 8x RTX A6000 | from $3.20 | Rent |
| FLUX.2-dev Black Forest Labs · Image editing | 32.2B | 136 GB | 4x RTX A6000 | from $1.60 | Rent |
| Llama-3.2-3B-Instruct Meta · Text generation | 3.2B | 8 GB | RTX 4060 Ti | from $0.14 | Rent |
| NVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GB NVIDIA · Vision and text | 24.6B | 29 GB | 2x RTX 5060 Ti | from $0.30 | Rent |
| Qwen3.6-35B-A3B Alibaba Qwen · Vision and text | 36B | 87 GB | 2x RTX A6000 | from $0.84 | Rent |
| Qwen3.5-4B Alibaba Qwen · Vision and text | 4.7B | 12 GB | RTX 5060 Ti | from $0.15 | Rent |
| LFM3-Audio-2B Liquid AI · Multimodal | 2B | 5 GB | GTX 1660 Ti | from $0.10 | Rent |
| gpt-oss-120b OpenAI · Text generation | 116.8B | 79 GB | A100 80GB PCIe | from $0.81 | Rent |
| FLUX.2-klein-9B Black Forest Labs · Image editing | 9.1B | 42 GB | 2x RTX 3090 | from $0.36 | Rent |
| gemma-4-26B-A4B-it Google · Vision and text | 25.8B | 62 GB | A100 80GB PCIe | from $0.81 | Rent |
| Qwen3.5-0.8B Alibaba Qwen · Vision and text | 0.9B | 3 GB | GTX 1660 Ti | from $0.10 | Rent |
| medgemma-1.5-4b-it Google · Vision and text | 4.3B | 11 GB | RTX 5060 Ti | from $0.15 | Rent |
| Llama-3.1-8B Meta · Text generation | 8B | 20 GB | RTX 3090 | from $0.16 | Rent |
| granite-4.1-8b-llmon-preview IBM · Text generation | 8.8B | 22 GB | RTX 3090 | from $0.16 | Rent |
| paligemma-3b-pt-224 Google · Vision and text | 2.9B | 15 GB | RTX 5060 Ti | from $0.15 | Rent |
| granite-4.1-3b-llmon-preview IBM · Text generation | 3.7B | 9 GB | RTX 5060 Ti | from $0.15 | Rent |
| Qwen3.8-2.4T-A95B Alibaba Qwen · Text generation | 2.4T | 5871 GB | A multi-node cluster | - | Request |
| gpt-oss-20b OpenAI · Text generation | 20.9B | 17 GB | RTX 3090 | from $0.16 | Rent |
| gemma-3-4b-it Google · Vision and text | 4.3B | 11 GB | RTX 5060 Ti | from $0.15 | Rent |
| gemma-4-E2B-it Google · Multimodal | 5.1B | 13 GB | RTX 5060 Ti | from $0.15 | Rent |
| gemma-4-12B Google · Multimodal | 12B | 29 GB | 2x RTX 5060 Ti | from $0.30 | Rent |
| Qwen3-0.6B Alibaba Qwen · Text generation | 0.8B | 2 GB | GTX 1660 Ti | from $0.10 | Rent |
| WeVisOne-VL-4B Tencent · Vision and text | 4.8B | 12 GB | RTX 5060 Ti | from $0.15 | Rent |
| FLUX.2-klein-4B Black Forest Labs · Image editing | 3.9B | 20 GB | RTX 3090 | from $0.16 | Rent |
| LFM3-VL-2B Liquid AI · Vision and text | 1.9B | 5 GB | GTX 1660 Ti | from $0.10 | Rent |
| Cosmos-Reason2-2B NVIDIA · Vision and text | 2.4B | 6 GB | GTX 1660 Ti | from $0.10 | Rent |
| Qwen3-Coder-30B-A3B-Instruct Alibaba Qwen · Text generation | 30.5B | 74 GB | A100 80GB PCIe | from $0.81 | Rent |
| Meta-Llama-3-8B Meta · Text generation | 8B | 20 GB | RTX 3090 | from $0.16 | Rent |
| Llama-3.2-3B Meta · Text generation | 3.2B | 8 GB | RTX 4060 Ti | from $0.14 | Rent |
| MiMo-V2.6-Distill-Qwen-9B Xiaomi · Vision and text | 9.4B | 23 GB | RTX 3090 | from $0.16 | Rent |
| Hy4-preview Tencent · Text generation | 780B | 1872 GB | 8x B300 SXM6 | from $32.00 | Rent |
| medgemma-4b-it Google · Vision and text | 4.3B | 11 GB | RTX 5060 Ti | from $0.15 | Rent |
| LFM2.5-24B-A2B Liquid AI · Text generation | 24B | 58 GB | A100 80GB PCIe | from $0.81 | Rent |
| Llama-3.2-1B Meta · Text generation | 1.2B | 3 GB | GTX 1660 Ti | from $0.10 | Rent |
| Ling-3.0-tiny Ant Group · Text generation | 7.9B | 19 GB | RTX 3090 | from $0.16 | Rent |
| granite-4.2-8b IBM · Text generation | 8.8B | 22 GB | RTX 3090 | from $0.16 | Rent |
| gemma-4-E4B-it Google · Multimodal | 8B | 20 GB | RTX 3090 | from $0.16 | Rent |
| gemma-4-12B-it Google · Multimodal | 12B | 29 GB | 2x RTX 5060 Ti | from $0.30 | Rent |
| d1-3B Liquid AI · Vision and text | 3.1B | 8 GB | RTX 4060 Ti | from $0.14 | Rent |
| MiMo-V2.6-Pro-RL Xiaomi · Text generation | 1T | 680 GB | 8x RTX PRO 6000 Workstation | from $15.92 | Rent |
| DeepSeek-V4-Flash-0731 DeepSeek · Text generation | 304.2B | 201 GB | 8x RTX A6000 | from $3.20 | Rent |
| DeepSeek-R1 DeepSeek · Text generation | 684.5B | 827 GB | 8x H200 | from $28.00 | Rent |
| DeepSeek-R1-Distill-Qwen-7B DeepSeek · Text generation | 7.6B | 19 GB | RTX 3090 | from $0.16 | Rent |
| Llama-3.3-70B-Instruct Meta · Text generation | 70.6B | 170 GB | 4x RTX A6000 | from $1.60 | Rent |
| Ling-3.1-flash Ant Group · Text generation | 572.1B | 689 GB | 8x RTX PRO 6000 Workstation | from $15.92 | Rent |
| Llama-3.2-1B-Instruct Meta · Text generation | 1.2B | 3 GB | GTX 1660 Ti | from $0.10 | Rent |
| MiMo-V2.6-Flash-RL Xiaomi · Text generation | 310.8B | 208 GB | 8x RTX A6000 | from $3.20 | Rent |
| FLUX.1-Kontext-dev Black Forest Labs · Image editing | 11.9B | 41 GB | 2x RTX 3090 | from $0.36 | Rent |
| Llama-2-7b-chat-hf Meta · Text generation | 6.7B | 17 GB | RTX 3090 | from $0.16 | Rent |
| DeepSeek-V4-Pro-0813 DeepSeek · Text generation | 1.7T | 1072 GB | 8x H200 | from $28.00 | Rent |
| DeepSeek-V3.2 DeepSeek · Text generation | 685.4B | 828 GB | 8x H200 | from $28.00 | Rent |
| GLM-4.7-Flash Z.ai · Text generation | 31.2B | 75 GB | A100 80GB PCIe | from $0.81 | Rent |
| MiniMax-M3 MiniMax · Vision and text | 427B | 1026 GB | 8x H200 | from $28.00 | Rent |
| SingVL_S1_4B Ant Group · Vision and text | 4.5B | 11 GB | RTX 5060 Ti | from $0.15 | Rent |
| HunyuanImage-3.0 Tencent · Image generation | 83B | 203 GB | 8x RTX A6000 | from $3.20 | Rent |
| WeVisOne-VL-2B Tencent · Vision and text | 2.4B | 6 GB | GTX 1660 Ti | from $0.10 | Rent |
| DeepSeek-R1-Distill-Qwen-1.5B DeepSeek · Text generation | 1.8B | 5 GB | GTX 1660 Ti | from $0.10 | Rent |
| DeepSeek-V4-Flash-Vision-Exp DeepSeek · Vision and text | 304.6B | 202 GB | 8x RTX A6000 | from $3.20 | Rent |
| GLM-5.2 Z.ai · Text generation | 753.3B | 1809 GB | 8x B300 SXM6 | from $32.00 | Rent |
| Llama-4-Maverick-17B-128E-Instruct Meta · Vision and text | 401.6B | 964 GB | 8x H200 | from $28.00 | Rent |
| deepseek-coder-6.7b-instruct DeepSeek · Text generation | 6.7B | 17 GB | RTX 3090 | from $0.16 | Rent |
| DeepSeek-R1-0528-Qwen3-8B DeepSeek · Text generation | 8.2B | 20 GB | RTX 3090 | from $0.16 | Rent |
| Kimi-Linear-48B-A3B-Instruct Moonshot AI · Text generation | 49.1B | 118 GB | 4x RTX A6000 | from $1.60 | Rent |
| Llama-3.1-405B Meta · Text generation | 405.9B | 975 GB | 8x H200 | from $28.00 | Rent |
| Llama-4-Scout-17B-16E-Instruct Meta · Vision and text | 108.6B | 261 GB | 8x RTX A6000 | from $3.20 | Rent |
| granite-4.2-3b IBM · Text generation | 3.7B | 9 GB | RTX 5060 Ti | from $0.15 | Rent |
| granite-4.2-30b IBM · Text generation | 29.3B | 71 GB | A100 80GB PCIe | from $0.81 | Rent |
| Kimi-VL-A3B-Thinking-2506 Moonshot AI · Vision and text | 16.4B | 40 GB | 2x RTX 3090 | from $0.36 | Rent |
| Kimi-K2-Thinking Moonshot AI · Text generation | 1T | 714 GB | 8x RTX PRO 6000 Workstation | from $15.92 | Rent |
| Kimi-K2.6 Moonshot AI · Vision and text | 1T | 715 GB | 8x RTX PRO 6000 Workstation | from $15.92 | Rent |
| Kimi-K2.7-Code Moonshot AI · Vision and text | 1T | 715 GB | 8x RTX PRO 6000 Workstation | from $15.92 | Rent |
| MiniMax-M2.5 MiniMax · Text generation | 228.7B | 277 GB | 8x RTX A6000 | from $3.20 | Rent |
| Mistral-7B-v0.1 Mistral AI · Text generation | 7.2B | 18 GB | RTX 3090 | from $0.16 | Rent |
| Molmo2-4B Ai2 · Vision and text | 4.9B | 24 GB | RTX 3090 | from $0.16 | Rent |
| Ling-3.0-flash Ant Group · Text generation | 127.5B | 306 GB | 8x RTX A6000 | from $3.20 | Rent |
| granite-docling-258M IBM · Vision and text | 0.3B | 1 GB | GTX 1660 Ti | from $0.10 | Rent |
| Moonlight-16B-A3B-Instruct Moonshot AI · Text generation | 16B | 39 GB | 2x RTX 3090 | from $0.36 | Rent |
| Kimi-VL-A3B-Instruct Moonshot AI · Vision and text | 16.4B | 40 GB | 2x RTX 3090 | from $0.36 | Rent |
| Kimi-Dev-72B Moonshot AI · Text generation | 72.7B | 175 GB | 4x RTX A6000 | from $1.60 | Rent |
| Kimi-K2.5 Moonshot AI · Vision and text | 1T | 715 GB | 8x RTX PRO 6000 Workstation | from $15.92 | Rent |
| codegeex4-all-9b Z.ai · Text generation | 9.4B | 23 GB | RTX 3090 | from $0.16 | Rent |
| GLM-4.5-Air Z.ai · Text generation | 110.5B | 266 GB | 8x RTX A6000 | from $3.20 | Rent |
| GLM-4.6V-Flash Z.ai · Vision and text | 10.3B | 25 GB | 2x RTX 5060 Ti | from $0.30 | Rent |
| MiniMax-M2.7 MiniMax · Text generation | 228.7B | 277 GB | 8x RTX A6000 | from $3.20 | Rent |
| OLMoE-1B-7B-0125 Ai2 · Text generation | 6.9B | 34 GB | 2x RTX 3090 | from $0.36 | Rent |
| OLMo-2-0325-32B-Instruct Ai2 · Text generation | 32.2B | 78 GB | A100 80GB PCIe | from $0.81 | Rent |
| Olmo-3-32B-Think Ai2 · Text generation | 32.2B | 78 GB | A100 80GB PCIe | from $0.81 | Rent |
| Olmo-3-7B-Instruct Ai2 · Text generation | 7.3B | 18 GB | RTX 3090 | from $0.16 | Rent |
| Olmo-3.1-32B-Think Ai2 · Text generation | 32.2B | 78 GB | A100 80GB PCIe | from $0.81 | Rent |
| MiMo-Audio-7B-Instruct Xiaomi · Multimodal | 8B | 20 GB | RTX 3090 | from $0.16 | Rent |
| MiMo-V2.6-Pro-MOPD Xiaomi · Text generation | 1T | 680 GB | 8x RTX PRO 6000 Workstation | from $15.92 | Rent |
| MiMo-V2.6-Flash-MOPD Xiaomi · Text generation | 310.8B | 208 GB | 8x RTX A6000 | from $3.20 | Rent |
| granite-3.1-3b-a800m-instruct IBM · Text generation | 3.3B | 8 GB | RTX 4060 Ti | from $0.14 | Rent |
| granite-3.3-2b-instruct IBM · Text generation | 2.5B | 7 GB | RTX 4060 Ti | from $0.14 | Rent |
| Step-Audio-2-mini StepFun · Multimodal | 8.3B | 20 GB | RTX 3090 | from $0.16 | Rent |
| ERNIE-Image-Turbo Baidu · Image generation | 8B | 38 GB | 2x RTX 3090 | from $0.36 | Rent |
| Moonlight-16B-A3B Moonshot AI · Text generation | 16B | 39 GB | 2x RTX 3090 | from $0.36 | Rent |
| MiniMax-Text-01 MiniMax · Text generation | 456.1B | 1098 GB | 8x H200 | from $28.00 | Rent |
| MiniMax-VL-01 MiniMax · Vision and text | 456.4B | 1099 GB | 8x H200 | from $28.00 | Rent |
| SynLogic-32B MiniMax · Text generation | 32.8B | 79 GB | A100 80GB PCIe | from $0.81 | Rent |
The list is the trending models from each lab's own Hugging Face account, refreshed every few hours. VRAM is the released weights plus about 20 % for activations and KV cache; long contexts and big batches need more.
·
Docs·
Blog·
For AI agents (CLI install, agent skill, headless signup): llms-full.txt