Best LLMs for 64 GB VRAM
64 GBFor example M1 Max, M5 Pro, RTX PRO 5000 Blackwell (72 GB)
Models that fit148entirely in memory
Best overall98 Qwen 2.5 72B Instruct
Largest that fits48.3 GB Qwen3-Coder-Next (80B-A3B MoE)
Best at 64 GB, by task
Score is the task-specific quality score on a 0–100 scale; speed is estimated for the M1 Max.
The strongest that fit
Top 15, by logic and coding
| Model | Memory · marker = 64 GB | VRAM | Speed on M1 Max (est.) | Score |
|---|---|---|---|---|
| Qwen 2.5 72B Instruct Qwen 2.5 Family · Q4_K_M | 43.5 GB | ~5 tok/s | 98 | |
| Qwen 2.5 Coder 32B Qwen 2.5 Family · Q4_K_M | 19.3 GB | ~10 tok/s | 97 | |
| Llama 3.3 70B Instruct Llama 3.3 · Q2_K_XS (Tight) | 20.2 GB | ~10 tok/s | 97 | |
| Qwen 3 32B Qwen 3 · Q4_K_M | 19.8 GB | ~10 tok/s | 97 | |
| DeepSeek R1 Distill Qwen 32B DeepSeek R1 · Q4_K_M | 19.3 GB | ~10 tok/s | 96 | |
| Qwen 2.5 14B Instruct Qwen 2.5 Family · Q4_K_M | 8.5 GB | ~22 tok/s | 95 | |
| Nemotron 70B Instruct Nemotron 70B · Q4_K_M | 42.6 GB | ~5 tok/s | 94 | |
| Qwen 3 14B Qwen 3 · Q4_K_M | 8.9 GB | ~21 tok/s | 94 | |
| Codestral 22B Codestral · Q4_K_M | 13.4 GB | ~15 tok/s | 93 | |
| Qwen 3.6 35B-A3B Qwen 3.6 · Q4_K_M | 21.1 GB | ~84 tok/s | 93 | |
| DeepSeek R1 Distill Qwen 14B DeepSeek R1 · Q4_K_M | 8.5 GB | ~22 tok/s | 92 | |
| Phi-4 (14B) Phi-4 Family · Q4_K_M | 8.5 GB | ~22 tok/s | 92 | |
| Gemma 4 31B Gemma 4 · Q4_K_M | 18.7 GB | ~10 tok/s | 92 | |
| Qwen 2.5 VL 72B Instruct Qwen 2.5 VL · Q4_K_M | 44.3 GB | ~5 tok/s | 92 | |
| Qwen3-Coder-Next (80B-A3B MoE) Qwen3-Coder · Q4_K_M | 48.3 GB | — | 92 |
What 80 GB unlocks
The strongest models that need the next tier up.
FAQ
What LLMs run well with 64 GB VRAM?
Qwen 2.5 Family, Qwen 2.5 Family, Llama 3.3, Qwen 3, DeepSeek R1 all fit in 64 GB VRAM.
Which GPUs have 64 GB VRAM?
Apple M1 Max, Apple M5 Pro, NVIDIA RTX PRO 5000 Blackwell (72 GB), NVIDIA A100 80GB (PCIe).
All tiers at a glance
| Memory | Models fit | Best overall | Largest that fits |
|---|---|---|---|
| 4 GB | 35 | Shieldstral 1.0 3B | Gemma 3 4B Instruct · 2.4 GB |
| 6 GB | 59 | Granite 3.0 8B Instruct | Qwen 2.5 VL 7B Instruct · 5 GB |
| 8 GB | 75 | Granite 3.0 8B Instruct | Falcon 3 10B Instruct · 6.2 GB |
| 10 GB | 87 | Qwen 2.5 14B Instruct | Qwen 3 14B · 8.9 GB |
| 12 GB | 89 | Qwen 2.5 14B Instruct | StarCoder 2 15B · 9.4 GB |
| 16 GB | 99 | Qwen 2.5 14B Instruct | Mistral Small 3 (24B) · 14.5 GB |
| 20 GB | 116 | Qwen 2.5 14B Instruct | Gemma 4 31B · 18.7 GB |
| 24 GB | 136 | Qwen 2.5 Coder 32B | Qwen 3.6 35B-A3B · 21.1 GB |
| 32 GB | 137 | Qwen 2.5 Coder 32B | Qwen 3.6 35B-A3B · 21.1 GB |
| 48 GB | 143 | Qwen 2.5 72B Instruct | Qwen 2.5 VL 72B Instruct · 44.3 GB |
| 64 GB | 148 | Qwen 2.5 72B Instruct | Qwen3-Coder-Next (80B-A3B MoE) · 48.3 GB |
| 80 GB | 158 | Qwen 2.5 72B Instruct | Qwen 3.5 122B-A10B · 73.7 GB |
or compare on Vast.ai from $0.77/hr (typical low · varies)
As an Amazon Associate we earn from qualifying purchases. Cloud GPU links are referral links — we may earn a commission at no extra cost to you.
Can-I-Run Pages Near 64 GB
- Nemotron 70B on Apple M1 Max
- Nemotron 70B on Apple M5 Pro
- Qwen 2.5 Family on Apple M1 Max
- Qwen 2.5 Family on Apple M5 Pro
- Llama 4 on Apple M1 Max
- Llama 4 on Apple M5 Pro
- Qwen 2.5 VL on Apple M1 Max
- Qwen 2.5 VL on Apple M5 Pro