The best open LLMs you can run on 16 GB of VRAM

Open LLMs that fit in 16 GB of VRAM at their default quant — enough for an RTX 4060 Ti 16 GB, 4070 Ti Super, or a 16 GB Mac. Ranked by the largest model that fits, with the honest $0-local cost. We confirm the fit; quality is yours to judge.

How this is ranked: Objective filter: 'best' = 'fits 16 GB.' Engine-computed VRAM, ordered by size. No subjective quality claim.

1. gpt oss safeguard 20b — openai, 21.5B · ~16 GB VRAM · $0.27/1M API est. · commercial OK
2. gpt-oss-20b — OpenAI, 21B · ~15 GB VRAM · $0.09/1M API · commercial OK
3. gpt oss 20b BF16 — unsloth, 20.9B · ~15 GB VRAM · $0.27/1M API est. · commercial OK
4. NVIDIA Nemotron 3 Nano 30B A3B NVFP4 — nvidia, 18.2B · ~13 GB VRAM · $0.25/1M API est. · non-commercial
5. Qwen3 30B A3B NVFP4 — RedHatAI, 17.5B · ~13 GB VRAM · $0.24/1M API est. · commercial OK
6. Qwen3 32B NVFP4 — nvidia, 17.2B · ~15 GB VRAM · $0.24/1M API est. · commercial OK
7. Param2 17B A2.4B Thinking — bharatgenai, 17.2B · ~12 GB VRAM · $0.24/1M API est. · non-commercial
8. LLaDA2.0 mini — inclusionAI, 16.3B · ~12 GB VRAM · $0.23/1M API est. · commercial OK
9. DeepSeek-Coder-V2-Lite Instruct — DeepSeek, 15.7B · ~11 GB VRAM · $0.23/1M API est. · commercial OK
10. DeepSeek V2 Lite Chat — deepseek-ai, 15.7B · ~15 GB VRAM · $0.23/1M API est. · non-commercial
11. DeepSeek V2 Lite — deepseek-ai, 15.7B · ~15 GB VRAM · $0.23/1M API est. · non-commercial
12. Qwen2.5 Coder 14B Instruct — Qwen, 14.8B · ~14 GB VRAM · $0.22/1M API est. · commercial OK
13. Qwen2.5 14B Instruct — Qwen, 14.8B · ~14 GB VRAM · $0.22/1M API est. · commercial OK
14. Qwen3 14B — Qwen, 14.8B · ~13 GB VRAM · $0.22/1M API est. · commercial OK
15. Qwen3 14B Base — Qwen, 14.8B · ~13 GB VRAM · $0.22/1M API est. · commercial OK
16. Qwen3 14B Instruct — OpenPipe, 14.8B · ~13 GB VRAM · $0.22/1M API est. · commercial OK
17. Qwen1.5 MoE A2.7B — Qwen, 14.3B · ~12 GB VRAM · $0.21/1M API est. · non-commercial
18. Phi-4 — Microsoft, 14B · ~13 GB VRAM · $0.21/1M API est. · commercial OK
19. Llama 2 13b chat hf — meta-llama, 13B · ~16 GB VRAM · $0.20/1M API est. · commercial OK
20. HarmBench Llama 2 13b cls — cais, 13B · ~11 GB VRAM · $0.20/1M API est. · commercial OK
21. NVIDIA Nemotron Nano 12B v2 — nvidia, 12.3B · ~15 GB VRAM · $0.20/1M API est. · non-commercial
22. MN 12B Mag Mell R1 — inflatebot, 12.2B · ~12 GB VRAM · $0.20/1M API est. · non-commercial
23. Gemma 3 12B — Google, 12B · ~13 GB VRAM · $0.20/1M API est. · commercial OK
24. Gemma 4 12B OBLITERATED — OBLITERATUS, 12B · ~14 GB VRAM · $0.20/1M API est. · commercial OK
25. Apertus 70B Instruct 2509 quantized.w4a16 — RedHatAI, 11.3B · ~14 GB VRAM · $0.19/1M API est. · commercial OK
26. Falcon3-10B Instruct — TII, 10B · ~10 GB VRAM · $0.18/1M API est. · commercial OK
27. Darwin 9B NEG — ansulev, 9.7B · ~12 GB VRAM · $0.18/1M API est. · commercial OK
28. SeeClick — cckevinn, 9.7B · ~12 GB VRAM · $0.18/1M API est. · non-commercial
29. gemma 2 9b — google, 9.2B · ~11 GB VRAM · $0.17/1M API est. · commercial OK
30. Gemma 2 9B Instruct — Google, 9B · ~9.0 GB VRAM · $0.06/1M API · commercial OK
31. NVIDIA Nemotron Nano 9B v2 — nvidia, 8.9B · ~10 GB VRAM · $0.17/1M API est. · non-commercial
32. NVIDIA Nemotron Nano 9B v2 Japanese — nvidia, 8.9B · ~10 GB VRAM · $0.17/1M API est. · non-commercial
33. internlm3 8b instruct — internlm, 8.8B · ~8.0 GB VRAM · $0.17/1M API est. · commercial OK
34. Nemotron Labs Diffusion 8B Base — nvidia, 8.5B · ~7.0 GB VRAM · $0.17/1M API est. · non-commercial
35. LFM2.5 8B A1B — LiquidAI, 8.5B · ~7.0 GB VRAM · $0.17/1M API est. · non-commercial
36. gemma 7b — google, 8.5B · ~11 GB VRAM · $0.17/1M API est. · commercial OK
37. Qwen3-8B — Alibaba, 8.2B · ~9.0 GB VRAM · $0.17/1M API est. · commercial OK
38. Qwen3 8B Base — Qwen, 8.2B · ~9.0 GB VRAM · $0.17/1M API est. · commercial OK
39. DeepSeek R1 0528 Qwen3 8B — deepseek-ai, 8.2B · ~9.0 GB VRAM · $0.17/1M API est. · commercial OK
40. Qwen3 14B NVFP4 — nvidia, 8.2B · ~9.0 GB VRAM · $0.17/1M API est. · commercial OK

Showing the top 40 of 237. See all →

More: all "best" lists · cost calculator · all models