What Can the AMD RX 9070 Run?

16 GB · compatibility with 168 AI models

Home › GPUs › AMD RX 9070

Short answer: the AMD RX 9070 (16 GB) can run 166 of 168 AI image, video, 3D and audio models we track. Best picks: MiniMax H3 FL2VA, MiniMax H3 Ref2VA, Qwen-Image 2.1, Z-Image Turbo, WAN 2.2 T2V A14B (MoE).

S Perfect	A Great	B Good	C Usable	D Barely	F Can't Run
13 models	64 models	71 models	10 models	8 models	2 models

Open the AMD RX 9070 in the checker

AMD RX 9070 specs for AI

Brand	AMD
Series	RX 9000
VRAM	16 GB
Architecture	rdna4
FP16 compute	28 TFLOPS
Memory bandwidth	576 GB/s
FP8 support	Yes
Released	2025
Launch price	$499

AI models on the AMD RX 9070

Each model uses the best format that fits in 16 GB. How ratings work.

Text to Image

Model	Rating	Runs as	VRAM used	Speed
Stable Diffusion 1.5	A Great	FP16	4 GB	Fast
Stable Diffusion 1.4	A Great	FP16	4 GB	Fast
Stable Diffusion 2.1	A Great	FP16	5 GB	Fast
Stable Diffusion 2.0	A Great	FP16	5 GB	Fast
PixArt-Alpha 600M	A Great	FP16	6 GB	Fast
SDXL Base 1.0	A Great	FP16	7 GB	Fast
SDXL Turbo	A Great	FP16	7 GB	Fast
Animagine XL 3.1	A Great	FP16	7 GB	Fast
Pony Diffusion v6 XL	A Great	FP16	7 GB	Fast
NoobAI XL	A Great	FP16	7 GB	Fast
Juggernaut XL Ragnarok	A Great	FP16	7 GB	Fast
SDXL Lightning 4-step	A Great	FP16	7 GB	Fast
DreamShaper XL	A Great	FP16	7 GB	Fast
Illustrious XL v1.0	A Great	FP16	7 GB	Fast
Playground v2.5 2.6B	A Great	FP16	7 GB	Fast
RealVisXL v5.0	A Great	FP16	7 GB	Fast
Epicrealism XL	A Great	FP16	7 GB	Fast
PixArt-Sigma 900M	A Great	FP16	7 GB	Fast
Stable Diffusion 3 Medium (2B)	A Great	FP16	7 GB	Fast
SDXL Hyper 1-step	A Great	FP16	7 GB	Fast
Stable Diffusion 3.5 Medium (2.5B)	A Great	FP16	8 GB	Fast
HunyuanDiT 1.5B	A Great	FP16	8 GB	Fast
Z-Image Turbo	B Good	FP16	10 GB	Fast
FLUX.2 Klein 9B	B Good	FP8	9 GB	Moderate
Chroma1 HD	B Good	FP8	9 GB	Moderate
MOSS-Image 5B	B Good	FP16	12 GB	Fast
Stable Diffusion 3.5 Large (8B)	B Good	FP8	10 GB	Moderate
Stable Diffusion 3.5 Large Turbo	B Good	FP8	10 GB	Moderate
Ideogram 4.0	B Good	FP8	10 GB	Moderate
Flux.1 Lite	B Good	FP16	12 GB	Fast
AuraFlow 0.3	B Good	FP16	12 GB	Fast
Flux.1 Dev 12B	B Good	FP8	12 GB	Moderate
Flux.2 Klein 4B	B Good	FP16	13 GB	Fast
Flux.1 Schnell 12B	B Good	FP8	12 GB	Moderate
FLUX.1 Krea Dev	B Good	FP8	12 GB	Moderate
HunyuanImage 3.0	B Good	FP8	12 GB	Moderate
Qwen-Image 2.1	B Good	FP16	14 GB	Moderate
Krea 2 Turbo	B Good	FP8	13 GB	Moderate
Z-Image Base	B Good	FP16	14 GB	Moderate
Ming-Image 0.1 Design	B Good	FP16	14 GB	Moderate
Kolors	B Good	FP16	14 GB	Moderate
Qwen-Image 2512 (20B)	B Good	GGUF Q4	12 GB	Moderate
Flux.2 Dev 32B	C Usable	GGUF Q3	14 GB	Slow
ERNIE-Image 8B	C Usable	FP16	16 GB	Moderate
HiDream-O1-Image 8.8B	C Usable	FP16	16 GB	Moderate
Seedream 3.0	D Barely	FP16	16 GB	Moderate
GLM-Image	D Barely	FP16	16 GB	Moderate

Image to Image

Model	Rating	Runs as	VRAM used	Speed
SD 1.5 img2img	A Great	FP16	4 GB	Fast
InstructPix2Pix	A Great	FP16	5 GB	Fast
SDXL img2img	A Great	FP16	7 GB	Fast
IP-Adapter SDXL	B Good	FP16	9 GB	Fast
IP-Adapter FaceID SDXL	B Good	FP16	10 GB	Fast
Mage-Flow 4B	B Good	FP16	10 GB	Fast
Boogu-Image Edit	B Good	FP8	11 GB	Moderate
Flux.1 Dev img2img 12B	B Good	FP8	12 GB	Moderate
Flux Kontext Dev	B Good	FP8	12 GB	Moderate
Qwen-Image-Edit 2511	B Good	GGUF Q4	12 GB	Moderate
LongCat-Image Edit	B Good	FP16	15 GB	Moderate

Inpainting

Model	Rating	Runs as	VRAM used	Speed
SD 1.5 Inpainting	A Great	FP16	4 GB	Fast
SDXL Inpainting	A Great	FP16	7 GB	Fast
PowerPaint v2.1	A Great	FP16	7 GB	Fast
BrushNet	A Great	FP16	8 GB	Fast
Flux.1 Fill Dev	B Good	FP8	12 GB	Moderate

ControlNet

Model	Rating	Runs as	VRAM used	Speed
ControlNet SD 1.5	A Great	FP16	5 GB	Fast
T2I-Adapter SDXL	A Great	FP16	8 GB	Fast
ControlNet SDXL	B Good	FP16	9 GB	Fast
ControlNet Flux	B Good	FP8	14 GB	Moderate

Video

Model	Rating	Runs as	VRAM used	Speed
WAN 2.1 T2V 1.3B	A Great	FP16	6 GB	Fast
AnimateDiff Lightning	A Great	FP16	6 GB	Fast
AnimateDiff v3 (SD 1.5)	A Great	FP16	6 GB	Fast
ZeroScope v2 576w	A Great	FP16	6 GB	Fast
AnimateDiff V2V	A Great	FP16	6 GB	Fast
AnimateDiff I2V	A Great	FP16	6 GB	Fast
Stable Video Diffusion 1.1	A Great	FP16	8 GB	Fast
CogVideoX 2B	A Great	FP16	8 GB	Fast
ModelScope T2V	A Great	FP16	8 GB	Fast
TokenFlow	A Great	FP16	8 GB	Fast
Latte	A Great	FP16	8 GB	Fast
SEINE	A Great	FP16	8 GB	Fast
Kandinsky 6.0 Video Lite 3B	B Good	FP16	9 GB	Fast
LTX-Video 0.9 (2B)	B Good	FP16	10 GB	Fast
LTX-Video I2V	B Good	FP16	10 GB	Fast
Stable Video Diffusion XT 1.1 (25 frames)	B Good	FP16	10 GB	Fast
DynamiCrafter 512	B Good	FP16	10 GB	Fast
WAN 2.2 TI2V 5B	B Good	FP16	12 GB	Fast
AnimateDiff SDXL	B Good	FP16	12 GB	Fast
Ovi 1.1	B Good	FP8	12 GB	Moderate
Wan 2.2 S2V 14B	B Good	GGUF Q4	10 GB	Moderate
Wan-Animate-2 14B	B Good	GGUF Q4	10 GB	Moderate
SCAIL-2 14B	B Good	GGUF Q4	10 GB	Moderate
Wan-Dancer 14B	B Good	GGUF Q4	10 GB	Moderate
HuMo 17B	B Good	GGUF Q4	10 GB	Moderate
CogVideoX 5B	B Good	FP16	14 GB	Moderate
CogVideoX 5B I2V	B Good	FP16	14 GB	Moderate
SkyReels V1	B Good	FP16	14 GB	Moderate
SkyReels V1 I2V	B Good	FP16	14 GB	Moderate
Wan 2.2 Animate 14B	B Good	GGUF Q4	11 GB	Moderate
LTX-2 19B	B Good	GGUF Q4	11 GB	Moderate
MiniMax H3 FL2VA	B Good	GGUF Q4	12 GB	Moderate
MiniMax H3 Ref2VA	B Good	GGUF Q4	12 GB	Moderate
WAN 2.2 T2V A14B (MoE)	B Good	FP8	14 GB	Moderate
WAN 2.2 I2V A14B	B Good	FP8	14 GB	Moderate
WAN 2.1 T2V 14B	B Good	FP8	14 GB	Moderate
WAN 2.1 I2V 480P 14B	B Good	FP8	14 GB	Moderate
WAN 2.1 V2V	B Good	FP8	14 GB	Moderate
Open-Sora 2.0 (11B)	B Good	FP8	14 GB	Moderate
LongCat-Video 13.6B	B Good	FP8	14 GB	Moderate
LTX-2.5 22B	B Good	GGUF Q4	13 GB	Moderate
LTX-2.3 22B	B Good	GGUF Q4	13 GB	Moderate
SkyReels V3 R2V 14B	B Good	FP8	15 GB	Moderate
DynamiCrafter 1024	C Usable	Offload	12 GB	Very Slow
WAN 2.1 I2V 720P 14B	C Usable	FP8	16 GB	Moderate
Waver 1.0	C Usable	FP16	12 GB	Fast
LTX-Video 13B	C Usable	FP8	14 GB	Moderate
MAGI-1	C Usable	FP8	14 GB	Moderate
HunyuanVideo T2V v1.5	D Barely	FP8	16 GB	Moderate
HunyuanVideo I2V v1.5	D Barely	FP8	16 GB	Moderate
HunyuanVideo 13B T2V	D Barely	FP8	16 GB	Moderate
HunyuanVideo I2V	D Barely	FP8	16 GB	Moderate
FramePack I2V HY FP8	D Barely	FP8	16 GB	Moderate
Mochi 1 10B	D Barely	Offload	16 GB	Very Slow
MOVA	F Can't Run	—	needs 29 GB	—

Upscaler

Model	Rating	Runs as	VRAM used	Speed
4x-UltraSharp	S Perfect	FP32	0.3 GB	Fast
4x-Remacri	S Perfect	FP32	0.3 GB	Fast
SwinIR	S Perfect	FP32	0.5 GB	Fast
RealESRGAN x4plus	S Perfect	FP32	0.3 GB	Fast
HAT	S Perfect	FP32	0.5 GB	Fast
DAT	S Perfect	FP32	0.5 GB	Fast
ScuNET	S Perfect	FP32	0.3 GB	Fast
LDSR	A Great	FP16	4 GB	Fast
SeedVR2 3B	B Good	FP16	9 GB	Fast
SeedVR2 7B	C Usable	FP16	16 GB	Moderate

Face/Restore

Model	Rating	Runs as	VRAM used	Speed
GFPGAN v1.4	S Perfect	FP32	0.3 GB	Fast
CodeFormer v0.1	S Perfect	FP32	0.3 GB	Fast

VAE

Model	Rating	Runs as	VRAM used	Speed
SD 1.5 VAE ft-mse	S Perfect	FP32	0.2 GB	Fast
Flux VAE	S Perfect	FP32	0.3 GB	Fast
SDXL VAE	S Perfect	FP32	0.3 GB	Fast
TAESD / TAESDXL	S Perfect	FP32	0.1 GB	Fast

3D

Model	Rating	Runs as	VRAM used	Speed
TripoSR	A Great	FP16	4 GB	Fast
Shap-E	A Great	FP16	4 GB	Fast
InstantMesh	A Great	FP16	6 GB	Fast
Zero123++ v1.2	A Great	FP16	6 GB	Fast
LGM	A Great	FP16	8 GB	Fast
One-2-3-45	A Great	FP16	8 GB	Fast
Hunyuan3D 2.1	B Good	FP16	10 GB	Fast
SV3D	B Good	FP16	10 GB	Fast
TRELLIS.2 4B	B Good	FP8	12 GB	Moderate
Pixal3D	B Good	FP16	14 GB	Moderate
Hunyuan3D 2.0	C Usable	FP16	12 GB	Fast
SAM 3D Objects	F Can't Run	—	needs 32 GB	—

Audio

Model	Rating	Runs as	VRAM used	Speed
MusicGen Small (300M)	A Great	FP16	2 GB	Fast
XTTS v2	A Great	FP16	3 GB	Fast
Stable Audio 3 Small-SFX	A Great	FP16	3 GB	Fast
Whisper Large v3	A Great	FP16	4 GB	Fast
Chatterbox Multilingual	A Great	FP16	4 GB	Fast
MMAudio Large 44k	A Great	FP16	4 GB	Fast
Stable Audio Open	A Great	FP16	4 GB	Fast
Qwen3-TTS 1.7B	A Great	FP16	5 GB	Fast
MusicGen Medium (1.5B)	A Great	FP16	6 GB	Fast
IndexTTS 2.5	A Great	FP16	6 GB	Fast
VibeVoice 1.5B	A Great	FP16	7 GB	Fast
Stable Audio 3 Medium	A Great	FP16	7 GB	Fast
ACE-Step 1.5 Turbo	A Great	FP16	8 GB	Fast
Bark	A Great	FP16	8 GB	Fast
MusicGen Large (3.3B)	B Good	FP16	12 GB	Fast
Fish Audio S2 Pro 4B	B Good	GGUF Q4	7 GB	Moderate
YuE2 3B	B Good	FP8	12 GB	Moderate
MiniMax Music 3	B Good	FP8	12 GB	Moderate

Frequently asked questions

How much VRAM does the AMD RX 9070 have?

The AMD RX 9070 has 16 GB of VRAM with 576 GB/s of memory bandwidth.

What AI models can the AMD RX 9070 run?

It can run 166 of the 168 AI models we track: 13 perfectly (S), 64 great (A), 71 good (B), 10 usable (C) and 8 barely (D). 2 need more VRAM than it has.

Can the AMD RX 9070 run MiniMax H3 FL2VA?

Yes, in GGUF Q4 (rating B, good).

Can the AMD RX 9070 run MiniMax H3 Ref2VA?