AMD · CDNA 3

Best AI Models for AMD Instinct MI300X (192 GB)

Runs everything up to 120B MoE well.

192 GB HBM35300 GB/s1307 TFLOPS FP16750 WCDNA 3 2023

What LLMs can the Instinct MI300X run?

#	Model	Speed	Quant · VRAM
1	Qwen3.8 27BAlibabaHow to run → Q4_K_M · 17.9 GB · 183 t/s	183 tok/s	Q4_K_M · 17.9 GB
2	Inkling SmallThinking Machines Q4_K_M · 162.7 GB · 115 t/s	115 tok/s	Q4_K_M · 162.7 GB
3	DeepSeek V4 FlashDeepSeek Q4_K_M · 175.1 GB · 88.7 t/s	88.7 tok/s	Q4_K_M · 175.1 GB
4	Qwen3.6 27BAlibaba Q4_K_M · 17.9 GB · 183 t/s	183 tok/s	Q4_K_M · 17.9 GB
5	NVIDIA Nemotron 3.5 Lightning 30B A3B BF16NVIDIA Q4_K_M · 19.3 GB · 92.1 t/s	92.1 tok/s	Q4_K_M · 19.3 GB

#	Model	Speed	Quant · VRAM
1	Qwen3.8 27BAlibabaHow to run → Q4_K_M · 17.9 GB · 183 t/s	183 tok/s	Q4_K_M · 17.9 GB
2	DeepSeek V4 FlashDeepSeek Q4_K_M · 175.1 GB · 88.7 t/s	88.7 tok/s	Q4_K_M · 175.1 GB
3	Gemma 4 31B ITGoogle Q4_K_M · 21.7 GB · 156 t/s	156 tok/s	Q4_K_M · 21.7 GB
4	Qwen3.6 35B A3BAlibaba Q4_K_M · 22.0 GB · 112 t/s	112 tok/s	Q4_K_M · 22.0 GB
5	NVIDIA Nemotron 3.5 Lightning 30B A3B BF16NVIDIA Q4_K_M · 19.3 GB · 92.1 t/s	92.1 tok/s	Q4_K_M · 19.3 GB

#	Model	Speed	Quant · VRAM
1	Qwen3.8 27BAlibabaHow to run → Q4_K_M · 17.9 GB · 183 t/s	183 tok/s	Q4_K_M · 17.9 GB
2	Inkling SmallThinking Machines Q4_K_M · 162.7 GB · 115 t/s	115 tok/s	Q4_K_M · 162.7 GB
3	Gemma 4 31B ITGoogle Q4_K_M · 21.7 GB · 156 t/s	156 tok/s	Q4_K_M · 21.7 GB
4	DeepSeek V4 FlashDeepSeek Q4_K_M · 175.1 GB · 88.7 t/s	88.7 tok/s	Q4_K_M · 175.1 GB
5	Qwen3.8 Flash NextAlibaba Q4_K_M · 109.0 GB · 47.0 t/s	47.0 tok/s	Q4_K_M · 109.0 GB

#	Model	Speed	Quant · VRAM
1	Qwen3.8 27BAlibabaHow to run → Q4_K_M · 17.9 GB · 183 t/s	183 tok/s	Q4_K_M · 17.9 GB
2	Inkling SmallThinking Machines Q4_K_M · 162.7 GB · 115 t/s	115 tok/s	Q4_K_M · 162.7 GB
3	Qwen3.6 27BAlibaba Q4_K_M · 17.9 GB · 183 t/s	183 tok/s	Q4_K_M · 17.9 GB
4	Qwen3.6 35B A3BAlibaba Q4_K_M · 22.0 GB · 112 t/s	112 tok/s	Q4_K_M · 22.0 GB
5	NVIDIA Nemotron 3.5 Lightning 30B A3B BF16NVIDIA Q4_K_M · 19.3 GB · 92.1 t/s	92.1 tok/s	Q4_K_M · 19.3 GB

#	Model	Speed	Quant · VRAM
1	Qwen3.8 27BAlibabaHow to run → Q4_K_M · 17.9 GB · 183 t/s	183 tok/s	Q4_K_M · 17.9 GB
2	Qwen3.6 27BAlibaba Q4_K_M · 17.9 GB · 183 t/s	183 tok/s	Q4_K_M · 17.9 GB
3	MiniMax M2.7MiniMax Q4_K_M · 139.6 GB · 69.0 t/s	69.0 tok/s	Q4_K_M · 139.6 GB
4	Gemma 4 31B ITGoogle Q4_K_M · 21.7 GB · 156 t/s	156 tok/s	Q4_K_M · 21.7 GB
5	NVIDIA Nemotron 3.5 Lightning 30B A3B BF16NVIDIA Q4_K_M · 19.3 GB · 92.1 t/s	92.1 tok/s	Q4_K_M · 19.3 GB

#	Model	Speed ↓	Quant · VRAM
1	Qwen3.5 4BAlibabaHow to run → Q4_K_M · 3.4 GB · 976 t/s	976 tok/s	Q4_K_M · 3.4 GB
2	Qwen3.5 9BAlibaba Q4_K_M · 6.6 GB · 500 t/s	500 tok/s	Q4_K_M · 6.6 GB
3	GPT OSS 20BOpenAI Q4_K_M · 13.1 GB · 178 t/s	178 tok/s	Q4_K_M · 13.1 GB
4	Gemma 4 31B ITGoogle Q4_K_M · 21.7 GB · 156 t/s	156 tok/s	Q4_K_M · 21.7 GB
5	NVIDIA Nemotron 3.5 Lightning 30B A3B BF16NVIDIA Q4_K_M · 19.3 GB · 92.1 t/s	92.1 tok/s	Q4_K_M · 19.3 GB

#	Model	Speed	Quant · VRAM
1	Qwen3.8 27BAlibabaHow to run → Q4_K_M · 17.9 GB · 183 t/s	183 tok/s	Q4_K_M · 17.9 GB
2	Inkling SmallThinking Machines Q4_K_M · 162.7 GB · 115 t/s	115 tok/s	Q4_K_M · 162.7 GB
3	Qwen3.6 27BAlibaba Q4_K_M · 17.9 GB · 183 t/s	183 tok/s	Q4_K_M · 17.9 GB
4	Gemma 4 31B ITGoogle Q4_K_M · 21.7 GB · 156 t/s	156 tok/s	Q4_K_M · 21.7 GB
5	JEV 27B VLautotrust Q4_K_M · 17.9 GB · 183 t/s	183 tok/s	Q4_K_M · 17.9 GB

See all 419 models on the Instinct MI300X →Best quant that fits · speeds estimated on this card

About the Instinct MI300X

The Instinct MI300X runs everything up to 120B MoE well. CDNA 3 generation, released 2023.

GenerationCDNA 3released 2023

Instinct MI300X strengths and weaknesses

Strengths

Runs 120B MoE models: GPT OSS 120B at 121 tok/s in 70.8 GB
Room for long context: Llama 3.3 70B Instruct at Q4 uses 51.3 GB, leaving 140.7 GB free
Reads long prompts fast: A 4K-token prompt on Qwen3.8 27B starts answering in 791 ms

Weaknesses

750 W under load: Plan a 1,500 W PSU; ≈$164 a year at 4 h a day

Instinct MI300X tokens per second by model size

Output speed depends on memory bandwidth (5300 GB/s).

9BQwen3.5 9B · Q4_K_M · 6.6 GB
500tok/s
14BQwen3 14B · Q4_K_M · 11.9 GB
335tok/s
27–31BQwen3.8 27B · Q4_K_M · 17.9 GB
183tok/s
35B MoEQwen3.6 35B A3B · Q4_K_M · 22.0 GB
112tok/s
70BLlama 3.3 70B Instruct · Q4_K_M · 51.3 GB
68.3tok/s
120B MoEGPT OSS 120B · Q4_K_M · 70.8 GB
121tok/s

Estimated speeds: Q4_K_M, 4K context, one user. How we estimate

All 1318 models on the Instinct MI300X

Featured only

Best quant that fits, ordered by llmrun Score. Fit grades how well it runs here, S best. How ranking works

LLM models compatible with AMD Instinct MI300X — ranked by Score, recency and Fit
Model	Quant	VRAM	Speed	Context	Fit
Baichuan 13B Base13BChat Q4_K_M·370.6 t/s	Q4_K_M	11.5 GB6%	370.6 t/s	—	S100, Runs great
Falcon 7B7.2BChat Q4_K_M·668.1 t/s	Q4_K_M	7.3 GB4%	668.1 t/s	—	S100, Runs great
Qwen3 30B A3B Base30.5BChat Q4_K_M·97.6 t/s	Q4_K_M	19.4 GB10%	97.6 t/s	33K	S91, Runs great
DeepSeek V3.1 Terminus684.5BChatFunctions IQ2_XXS·63.5 t/s	IQ2_XXS	189.7 GB99%	63.5 t/s	164K	C44, Usable
Deepseek Coder 7B Instruct V1.56.9BChatCode Q4_K_M·583.5 t/s	Q4_K_M	6.5 GB3%	583.5 t/s	4K	S100, Runs great
Phi 3 Vision 128k Instruct4.1BChatCodeVision BF16·338.3 t/s	BF16	15.0 GB8%	338.3 t/s	131K	S100, Runs great
Meta Llama 3 8B Instruct8.0BChat Q4_K_M·590.0 t/s	Q4_K_M	6.2 GB3%	590.0 t/s	8K	S100, Runs great
Llama 2 7B HF6.7BChat Q4_K_M·586.7 t/s	Q4_K_M	6.5 GB3%	586.7 t/s	4K	S100, Runs great
Qwen2.5 1.5B1.5BChatFunctions Q4_K_M·2484.4 t/s	Q4_K_M	1.7 GB1%	2484.4 t/s	131K	S100, Runs great
Llama 3.2 1B1.2BChat Q4_K_M·3878.0 t/s	Q4_K_M	2.3 GB1%	3878.0 t/s	—	S100, Runs great
SmolLM 360M362MChat Q4_K_M·5300.0 t/s	Q4_K_M	0.6 GB0%	5300.0 t/s	2K	S100, Runs great
SmolLM 360M Instruct362MChat Q4_K_M·5300.0 t/s	Q4_K_M	0.6 GB0%	5300.0 t/s	2K	S100, Runs great
SmolLM2 360M362MChat Q4_K_M·5300.0 t/s	Q4_K_M	0.8 GB0%	5300.0 t/s	8K	S100, Runs great
Qwen 14B Chat14.2BChat Q4_K_M·340.1 t/s	Q4_K_M	10.6 GB6%	340.1 t/s	8K	S100, Runs great
Aya 23 8B8.0BChat BF16·180.1 t/s	BF16	20.2 GB11%	180.1 t/s	—	S99, Runs great

Instinct MI300X specs, power and PSU

Specs

Architecture: CDNA 3
VRAM: 192 GB HBM3
Memory bandwidth: 5300.0 GB/s
FP16 (dense): 1307.40 TFLOPS
Stream processors: 19,456
TDP: 750 W
Released: 2023-12-06

Instinct MI300X FAQ

What is the largest model the Instinct MI300X can run?

MiMo V2.6 Flash RL (309B) is the largest at 4-bit or better: at Q4_K_M it uses 186.5 GB of 192 GB, and runs at about 80.4 tok/s.

Can the Instinct MI300X run Qwen3.8 27B?

Yes. Qwen3.8 27B fits at Q4_K_M, using 17.9 GB of 192 GB, and it is the best model that runs well on this card.

How many tokens per second does the Instinct MI300X give?

About 183 tok/s on Qwen3.8 27B at Q4_K_M, with the first token after 0.8 s on a 4K-token prompt.

What power supply does the Instinct MI300X need?

Plan a 1500 W PSU (2× the 750 W TDP).

Is 192 GB enough for coding and agents?

Yes for coding: Qwen3.8 27B, the best of the 19 coding models that fit, runs at about 183 tok/s.

Best coding LLMs 48 GB VRAM GPUs with 192 GB or more RTX 5090 RTX 4090 RTX 3090