readygpu

AMD Radeon RX 7900 XT vs AMD Radeon RX 9070

For running models locally, priced in the UK. At 8k context the first runs 52 models fully in memory, the second 37.

In short

The AMD Radeon RX 7900 XT has more memory for models: 20 GB against 16 GB. At 8k context the AMD Radeon RX 7900 XT runs 52 of the models we track fully in memory and the AMD Radeon RX 9070 37; only the AMD Radeon RX 7900 XT holds Qwen 3.8 27B. On the models both run fully in memory, the AMD Radeon RX 7900 XT is about 1.3× as fast as the AMD Radeon RX 9070 (estimated, median across models).

Updated . Speeds are estimates; how they are worked out.

AMD Radeon RX 7900 XT

£789

Amazon

Add to build
AMD Radeon RX 9070

Not priced here

Add to build

Specifications

SpecAMD Radeon RX 7900 XTAMD Radeon RX 9070
GPURadeon RX 7900 XTRadeon RX 9070
Memory20 GB GDDR616 GB GDDR6
Memory bus320-bit256-bit
Memory bandwidth800 GB/s640 GB/s
FP16 compute103 TFLOPS36.1 TFLOPS
SoftwareROCmROCm
Board power315 W220 W
PCIe generationGen 4Gen 5
PCIe lanesx16x16

What each runs

Estimated decode speed with the most precise quant that fits each one est.

ModelAMD Radeon RX 7900 XTAMD Radeon RX 9070
Qwen 3.8 27B24tok/sQ4_K_Moffload
Qwen 3.6 27B25tok/sQ4_K_Moffload
Qwen 3.5 27B25tok/sQ4_K_Moffload
Gemma 4 26B-A4B78tok/sQ5_K_Moffload
GLM 4.7 Flash76tok/sQ4_K_Moffload
Devstral Small 2 24B24tok/sQ5_K_Moffload
Qwen3 Coder 30B-A3B100tok/sQ4_K_Moffload
gpt-oss 20B100tok/sMXFP480tok/sMXFP4
Gemma 4 12B34tok/sQ8_027tok/sQ8_0
Qwen 3.5 9B45tok/sQ8_036tok/sQ8_0
Gemma 3 27B25tok/sQ4_K_Moffload
Mistral Small 3.2 24B24tok/sQ5_K_Moffload
Ministral 3 14B28tok/sQ8_028tok/sQ6_K
Qwen3 14B26tok/sQ8_026tok/sQ6_K
DeepSeek R1 0528 Qwen3 8B44tok/sQ8_036tok/sQ8_0
Qwen3 8B44tok/sQ8_036tok/sQ8_0
Phi-4 14B26tok/sQ8_026tok/sQ6_K
Qwen 3.5 4B92tok/sQ8_073tok/sQ8_0
Ministral 3 8B43tok/sQ8_035tok/sQ8_0
Granite 4.2 8B41tok/sQ8_033tok/sQ8_0
Gemma 4 E4B52tok/sQ8_042tok/sQ8_0
LFM2.5 8B-A1B180tok/sQ8_0144tok/sQ8_0
Llama 3.1 8B57tok/sQ6_K46tok/sQ6_K
Qwen2.5 0.5B697tok/sQ8_0557tok/sQ8_0
Qwen3.5 0.8B472tok/sQ8_0378tok/sQ8_0
Qwen3.5 2B283tok/sQ5_K_M226tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B207tok/sQ8_0165tok/sQ8_0
MiniCPM5 2B145tok/sQ8_0116tok/sQ8_0
LFM2.5 2.6B130tok/sQ8_0104tok/sQ8_0
LFM2.5 VL 3B130tok/sQ8_0104tok/sQ8_0
Spark X2.5 4B130tok/smlx-4bit104tok/smlx-4bit
LFM2.5 24B A2B119tok/sQ5_K_Moffload
granite vision 4.1 4b103tok/sQ8_082tok/sQ8_0
Ministral 3 3B Instruct 251297tok/sQ8_078tok/sQ8_0
granite 4.2 3b96tok/sQ8_077tok/sQ8_0
Mellum2.1 12B A2.5B Thinking94tok/sQ8_075tok/sQ8_0
FrogNano 4B 260989tok/sQ8_071tok/sQ8_0
Kimi VL A3B Thinking 250679tok/sQ8_097tok/sQ4_K_M
Kimi VL A3B75tok/smlx-8bit76tok/smlx-6bit
DeepSeek R1 Distill Qwen 7B51tok/sQ8_041tok/sQ8_0
GLM 4 9B 041451tok/smlx-5bit41tok/smlx-5bit
DeepSeek R1 Distill Llama 8B46tok/sQ8_037tok/sQ8_0
Ministral 3 8B Reasoning 251243tok/sQ8_035tok/sQ8_0
GLM 4.6V Flash43tok/sQ8_034tok/sQ8_0
GLM 4.1V 9B Thinking36tok/smlx-8bit29tok/smlx-8bit
Ministral 3 14B Reasoning 251228tok/sQ8_028tok/sQ6_K
DeepSeek R1 Distill Qwen 14B25tok/sQ8_026tok/sQ6_K
Muse Glimmer 30B25tok/sQ4_K_Moffload
Devstral Small 250724tok/sQ5_K_Moffload
Magistral Small 250624tok/sQ5_K_Moffload
Magistral Small 250724tok/sQ5_K_Moffload
Magistral Small 250924tok/sQ5_K_Moffload
Qwen 3.5 122B-A10Boffloadoffload
Gemma 4 31Boffloadoffload
Mistral Small 4 119Boffloadoffload
Qwen 3.6 35B-A3Boffloadoffload
GLM 4.5 Airoffloadwon't run
Qwen 3.5 35B-A3Boffloadoffload
Nemotron 3.5 Lightning 30B-A3Boffloadoffload
Kimi Linear 48B-A3Boffloadoffload
Qwen3 32Boffloadoffload
DeepSeek R1 Distill Qwen 32Boffloadoffload
Llama 3.3 70Boffloadwon't run
granite 4.2 30boffloadoffload
Qwen3.8 27B OBLITERATEDoffloadoffload
DeepSeek R1 Distill Llama 70Boffloadwon't run
DeepSeek V4 Flashwon't runwon't run
GLM 5.3 Flashwon't runwon't run
Qwen 3.8 Flash Nextwon't runwon't run
MiniMax M2.7won't runwon't run
Nemotron 3 Super 120B-A12Bwon't runwon't run
Qwen3 Coder Nextwon't runwon't run
gpt-oss 120Bwon't runwon't run
Qwen3 Next 80B-A3Bwon't runwon't run
DeepSeek R1won't runwon't run
DeepSeek R1 0528won't runwon't run
DeepSeek R1 Zerowon't runwon't run
DeepSeek V3won't runwon't run
DeepSeek V3 0324won't runwon't run
DeepSeek V3.1won't runwon't run
DeepSeek V3.1 Terminuswon't runwon't run
DeepSeek V3.2won't runwon't run
DeepSeek V4 Flash 0731won't runwon't run
DeepSeek V4 Flash Vision Expwon't runwon't run
Devstral 2 123B Instruct 2512won't runwon't run
GLM 4.5won't runwon't run
GLM 4.5Vwon't runwon't run
GLM 4.6won't runwon't run
GLM 4.6Vwon't runwon't run
GLM 4.7won't runwon't run
GLM 5won't runwon't run
GLM 5.1won't runwon't run
GLM 5.2won't runwon't run
GLM 5.3won't runwon't run
Kimi Dev 72Bwon't runwon't run
Kimi K2won't runwon't run
Kimi K2 Instruct 0905won't runwon't run
Kimi K2 Thinkingwon't runwon't run
Kimi K2.5won't runwon't run
MiniMax M2won't runwon't run
MiniMax M2.1won't runwon't run
MiniMax M2.5won't runwon't run
MiniMax M3won't runwon't run
Mistral Medium 3.5 128Bwon't runwon't run

Questions

Which is better for local AI, the AMD Radeon RX 7900 XT or the AMD Radeon RX 9070?
Choose the AMD Radeon RX 7900 XT to run bigger models. Choose the AMD Radeon RX 7900 XT for speed.
Which can run bigger models, the AMD Radeon RX 7900 XT or the AMD Radeon RX 9070?
At 8k context the AMD Radeon RX 7900 XT runs 52 of the models we track fully in memory and the AMD Radeon RX 9070 37; only the AMD Radeon RX 7900 XT holds Qwen 3.8 27B.
Which is faster for local AI, the AMD Radeon RX 7900 XT or the AMD Radeon RX 9070?
On the models both run fully in memory, the AMD Radeon RX 7900 XT is about 1.3× as fast as the AMD Radeon RX 9070 (estimated, median across models).

More comparisons

More hardware for local AI