readygpu

NVIDIA RTX PRO 4000 Blackwell vs AMD Radeon RX 7900 XTX

For running models locally, priced in the US. At 8k context the first runs 60 models fully in memory, the second 60.

In short

Both have 24 GB of memory for models; the NVIDIA RTX PRO 4000 Blackwell has 672 GB/s of bandwidth and the AMD Radeon RX 7900 XTX 960 GB/s. At 8k context both run the same 60 models we track fully in memory. On the models both run fully in memory, the AMD Radeon RX 7900 XTX is about 1.3× as fast as the NVIDIA RTX PRO 4000 Blackwell (estimated, median across models).

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

NVIDIA RTX PRO 4000 Blackwell

Not priced here

Add to build
AMD Radeon RX 7900 XTX

$1,250

Amazon

Add to build

Specifications

SpecNVIDIA RTX PRO 4000 BlackwellAMD Radeon RX 7900 XTX
GPURTX PRO 4000 BlackwellRadeon RX 7900 XTX
Memory24 GB GDDR724 GB GDDR6
Memory bus192-bit384-bit
Memory bandwidth672 GB/s960 GB/s
FP16 compute46.9 TFLOPS123 TFLOPS
SoftwareCUDAROCm
Board power145 W355 W
PCIe generationGen 5Gen 4
PCIe lanesx16x16

What each runs

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

ModelNVIDIA RTX PRO 4000 BlackwellAMD Radeon RX 7900 XTX
Qwen 3.8 27B18tok/sUD-Q6_K23tok/sUD-Q6_K
Qwen 3.6 27B18tok/sQ6_K23tok/sQ6_K
Gemma 4 31B18tok/sQ5_K_M23tok/sQ5_K_M
Qwen 3.6 35B-A3B120tok/sQ4_K_M152tok/sQ4_K_M
Qwen 3.5 27B18tok/sQ6_K23tok/sQ6_K
Gemma 4 26B-A4B63tok/sUD-Q6_K80tok/sUD-Q6_K
Qwen 3.5 35B-A3B121tok/sQ4_K_M153tok/sQ4_K_M
GLM 4.7 Flash63tok/sQ5_K_M79tok/sQ5_K_M
Devstral Small 2 24B20tok/sQ6_K26tok/sQ6_K
Nemotron 3.5 Lightning 30B-A3B109tok/sMXFP4138tok/sMXFP4
gpt-oss 20B95tok/sMXFP4120tok/sMXFP4
Qwen3 Coder 30B-A3B85tok/sQ5_K_M108tok/sQ5_K_M
Qwen3 32B19tok/sQ4_K_M24tok/sQ4_K_M
Gemma 4 12B32tok/sQ8_040tok/sQ8_0
Qwen 3.5 9B42tok/sQ8_054tok/sQ8_0
Mistral Small 3.2 24B20tok/sQ6_K26tok/sQ6_K
DeepSeek R1 Distill Qwen 32B19tok/sQ4_K_M24tok/sQ4_K_M
Gemma 3 27B18tok/sQ6_K23tok/sQ6_K
Ministral 3 14B27tok/sQ8_034tok/sQ8_0
Qwen3 14B24tok/sQ8_031tok/sQ8_0
DeepSeek R1 0528 Qwen3 8B42tok/sQ8_053tok/sQ8_0
Qwen3 8B42tok/sQ8_053tok/sQ8_0
Phi-4 14B24tok/sQ8_031tok/sQ8_0
Qwen 3.5 4B87tok/sQ8_0110tok/sQ8_0
Ministral 3 8B41tok/sQ8_052tok/sQ8_0
Granite 4.2 8B39tok/sQ8_049tok/sQ8_0
Gemma 4 E4B49tok/sQ8_062tok/sQ8_0
LFM2.5 8B-A1B170tok/sQ8_0216tok/sQ8_0
Llama 3.1 8B54tok/sQ6_K69tok/sQ6_K
Qwen2.5 0.5B660tok/sQ8_0836tok/sQ8_0
Qwen3.5 0.8B447tok/sQ8_0567tok/sQ8_0
Qwen3.5 2B268tok/sQ5_K_M340tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B196tok/sQ8_0248tok/sQ8_0
MiniCPM5 2B137tok/sQ8_0174tok/sQ8_0
LFM2.5 2.6B123tok/sQ8_0156tok/sQ8_0
LFM2.5 VL 3B123tok/sQ8_0156tok/sQ8_0
Spark X2.5 4B123tok/smlx-4bit155tok/smlx-4bit
LFM2.5 24B A2B103tok/sQ6_K130tok/sQ6_K
granite vision 4.1 4b97tok/sQ8_0123tok/sQ8_0
Ministral 3 3B Instruct 251292tok/sQ8_0117tok/sQ8_0
granite 4.2 3b91tok/sQ8_0116tok/sQ8_0
Mellum2.1 12B A2.5B Thinking89tok/sQ8_0112tok/sQ8_0
FrogNano 4B 260984tok/sQ8_0107tok/sQ8_0
Kimi VL A3B Thinking 250674tok/sQ8_094tok/sQ8_0
Kimi VL A3B71tok/smlx-8bit90tok/smlx-8bit
DeepSeek R1 Distill Qwen 7B49tok/sQ8_062tok/sQ8_0
GLM 4 9B 041449tok/smlx-5bit62tok/smlx-5bit
DeepSeek R1 Distill Llama 8B43tok/sQ8_055tok/sQ8_0
Ministral 3 8B Reasoning 251241tok/sQ8_052tok/sQ8_0
GLM 4.6V Flash40tok/sQ8_051tok/sQ8_0
GLM 4.1V 9B Thinking34tok/smlx-8bit44tok/smlx-8bit
Ministral 3 14B Reasoning 251227tok/sQ8_034tok/sQ8_0
DeepSeek R1 Distill Qwen 14B24tok/sQ8_031tok/sQ8_0
Devstral Small 250723tok/sQ5_K_M29tok/sQ5_K_M
Magistral Small 250723tok/sQ5_K_M29tok/sQ5_K_M
Magistral Small 250923tok/sQ5_K_M29tok/sQ5_K_M
Magistral Small 250620tok/sQ6_K26tok/sQ6_K
Qwen3.8 27B OBLITERATED20tok/smlx-4bit25tok/smlx-4bit
granite 4.2 30b18tok/sQ5_K_M23tok/sQ5_K_M
Muse Glimmer 30B18tok/sQ6_K22tok/sQ6_K
Qwen 3.5 122B-A10Boffloadoffload
Qwen3 Coder Nextoffloadoffload
Mistral Small 4 119Boffloadoffload
GLM 4.5 Airoffloadoffload
Qwen3 Next 80B-A3Boffloadoffload
Kimi Linear 48B-A3Boffloadoffload
Llama 3.3 70Boffloadoffload
DeepSeek R1 Distill Llama 70Boffloadoffload
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
gpt-oss 120Bwon'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 NVIDIA RTX PRO 4000 Blackwell or the AMD Radeon RX 7900 XTX?
Choose the AMD Radeon RX 7900 XTX for speed.
Which can run bigger models, the NVIDIA RTX PRO 4000 Blackwell or the AMD Radeon RX 7900 XTX?
At 8k context both run the same 60 models we track fully in memory.
Which is faster for local AI, the NVIDIA RTX PRO 4000 Blackwell or the AMD Radeon RX 7900 XTX?
On the models both run fully in memory, the AMD Radeon RX 7900 XTX is about 1.3× as fast as the NVIDIA RTX PRO 4000 Blackwell (estimated, median across models).

More comparisons

More hardware for local AI