GPU GUIDE · AMD
Best AI models for the
AMD Instinct MI210
The AMD Instinct MI210 has 64.0GB of VRAM. Below are the top 30 open-source AI models that fit, ranked by composite benchmark score. Each row shows the best quantization that fits your hardware.
VRAM
64.0GB
Brand
amd
Models that fit
30
Generation
Instinct
Top 30 models for the AMD Instinct MI210
01
77.9Qwen3.8 2.4T A95B
95.0BBest fit: Q4_K_M · 58.6GBCan I run it? →
02
69.0Qwen3.8 27B
27.0BBest fit: fp16 · 55.0GBCan I run it? →
03
48.3Qwen 3.6 27B
27.8BBest fit: fp16 · 56.6GBCan I run it? →
04
44.9Qwen3.5-27B
27.0BBest fit: fp16 · 55.0GBCan I run it? →
05
45.0Qwen 3 Next 80B A3B
80.0BBest fit: Q5_K_M · 57.8GBCan I run it? →
06
43.7Qwen 3.6 35B A3B
36.0BBest fit: Q8_0 · 39.3GBCan I run it? →
07
40.6Muse Glimmer 30B
30.0BBest fit: fp16 · 61.0GBCan I run it? →
08
37.7Qwen 3.5 35B A3B
36.0BBest fit: Q8_0 · 39.3GBCan I run it? →
09
37.0Gemma 4 31B
31.3BBest fit: fp16 · 63.6GBCan I run it? →
10
36.7Devstral 2
24.0BBest fit: fp16 · 49.0GBCan I run it? →
11
31.9Gemma 4 26B A4B
26.0BBest fit: fp16 · 53.0GBCan I run it? →
12
31.7Devstral Small 2
7.0BBest fit: f32 · 29.0GBCan I run it? →
13
26.7Ministral 3 14B
14.0BBest fit: f32 · 57.0GBCan I run it? →
14
25.9gemma 4 12B it
12.0BBest fit: f32 · 49.0GBCan I run it? →
15
25.0Ministral 3 8B
8.0BBest fit: f32 · 33.0GBCan I run it? →
16
25.0Gemma 4 E2B
5.1BBest fit: f32 · 21.4GBCan I run it? →
17
24.6Qwen 3.5 9B
9.7BBest fit: f32 · 39.8GBCan I run it? →
18
23.2Qwen 3.5 4B
4.7BBest fit: f32 · 19.8GBCan I run it? →
19
23.2Granite 4.2 8B
8.0BBest fit: fp16 · 17.0GBCan I run it? →
20
18.6Devstral Small 2 24B Instruct 2512
24.0BBest fit: fp16 · 49.0GBCan I run it? →
21
18.3Ministral 3 3B
3.0BBest fit: f32 · 13.0GBCan I run it? →
22
17.8Qwen3 Next 80B A3B Thinking
81.3BBest fit: Q4_K_M · 50.3GBCan I run it? →
23
15.2GPT-OSS 20B
20.0BBest fit: fp16 · 41.0GBCan I run it? →
24
15.0Nemotron 3 Nano Omni (free)
30.0BBest fit: fp16 · 61.0GBCan I run it? →
25
14.3Nemotron 3 Nano 30B A3B
30.0BBest fit: fp16 · 61.0GBCan I run it? →
26
14.3Qwen3 30B A3B Thinking 2507
30.5BBest fit: Q8_0 · 33.4GBCan I run it? →
27
13.1Qwen3 Next 80B A3B Instruct
81.3BBest fit: Q5_K_M · 58.7GBCan I run it? →
28
12.9Qwen3 Coder 30B A3B Instruct
30.5BBest fit: Q8_0 · 33.4GBCan I run it? →
29
12.7diffusiongemma 26B A4B it
25.8BBest fit: Q8_0 · 28.4GBCan I run it? →
30
12.5Qwen3 VL 30B A3B Thinking
31.1BBest fit: fp16 · 63.2GBCan I run it? →
Advertisement
FAQ — running AI on the AMD Instinct MI210
How many AI models can the AMD Instinct MI210 run?
With 64.0GB of VRAM, the AMD Instinct MI210 can run 30+ open-source models from our database, including Qwen3.8 2.4T A95B, Qwen3.8 27B, Qwen 3.6 27B.
What's the largest LLM I can run on a AMD Instinct MI210?
The biggest model that fits is approximately 95.0B. Larger models would need to be quantized further or won't fit at all.
Is 64.0GB of VRAM enough for local AI?
Yes — 64.0GB comfortably runs most popular open-source models including 30B-class LLMs at Q4_K_M.