Autor: Jakub Rusinowski · Ostatnia aktualizacja: 15 czerwca 2026
These are the strongest local models that fit entirely in 80 GB of VRAM, ranked by capability, with the quantization level and estimated tokens/sec needed to fit.
| Qwen 2.5 Family — Qwen 2.5 72B Instruct | Q4_K_M · 43 GB · ~6 tok/s on AMD Ryzen AI Max+ 395 |
| Kimi K2.5 — Kimi K2.5 32B Active | Q4_K_M · 45 GB · ~178 tok/s on AMD Ryzen AI Max+ 395 |
| Qwen 2.5 Family — Qwen 2.5 Coder 32B | Q4_K_M · 19.5 GB · ~13 tok/s on AMD Ryzen AI Max+ 395 |
| Llama 3.3 — Llama 3.3 70B Instruct | Q2_K_XS (Tight) · 26 GB · ~10 tok/s on AMD Ryzen AI Max+ 395 |
| Qwen 3 — Qwen 3 32B | Q4_K_M · 19.5 GB · ~13 tok/s on AMD Ryzen AI Max+ 395 |
| DeepSeek R1 — DeepSeek R1 Distill Qwen 32B | Q4_K_M · 19.5 GB · ~13 tok/s on AMD Ryzen AI Max+ 395 |
| Kimi K2.5 / K2.6 — Kimi K2.6 | Q4_K_M · 19 GB · ~13 tok/s on AMD Ryzen AI Max+ 395 |
| Qwen 2.5 Family — Qwen 2.5 14B Instruct | Q4_K_M · 9.5 GB · ~27 tok/s on AMD Ryzen AI Max+ 395 |
| Kimi K2.5 / K2.6 — Kimi K2.5 | Q4_K_M · 19 GB · ~13 tok/s on AMD Ryzen AI Max+ 395 |
| Nemotron 70B — Nemotron 70B Instruct | Q4_K_M · 39 GB · ~7 tok/s on AMD Ryzen AI Max+ 395 |
| Qwen 3.5 (Legacy Listing — Unverified) — Qwen 3.5 122B-A10B (MoE) | Q4_K_M · 13.5 GB · ~231 tok/s on AMD Ryzen AI Max+ 395 |
| Gemma 4 (Legacy Listing — Unverified) — Gemma 4 27B ⭐ | Q4_K_M · 14 GB · ~18 tok/s on AMD Ryzen AI Max+ 395 |
| Qwen 3 — Qwen 3 14B | Q4_K_M · 9.5 GB · ~27 tok/s on AMD Ryzen AI Max+ 395 |
| Qwen 3.5 (Legacy Listing — Unverified) — Qwen 3.5 72B | Q4_K_M · 42 GB · ~6 tok/s on AMD Ryzen AI Max+ 395 |
| Llama 4 — Llama 4 Maverick 17B | Q4_K_M · 24 GB · ~251 tok/s on AMD Ryzen AI Max+ 395 |
Qwen 2.5 Family, Kimi K2.5, Qwen 2.5 Family, Llama 3.3, Qwen 3 all fit in 80 GB VRAM.
AMD Ryzen AI Max+ 395, Apple M2 Max, Apple M4 Max, Apple M5 Max.