readygpu

NVIDIA RTX PRO 6000 Blackwell vs AMD Radeon RX 7900 XT

For running models locally, priced in the US. At 8k context the first runs 75 models fully in memory, the second 49.

In short

The NVIDIA RTX PRO 6000 Blackwell has more memory for models: 96 GB against 20 GB. At 8k context the NVIDIA RTX PRO 6000 Blackwell runs 75 of the models we track fully in memory and the AMD Radeon RX 7900 XT 49; only the NVIDIA RTX PRO 6000 Blackwell holds DeepSeek V4 Flash. On the models both run fully in memory, the NVIDIA RTX PRO 6000 Blackwell is about 2.5× as fast as the AMD Radeon RX 7900 XT (estimated, median across models). In the US the NVIDIA RTX PRO 6000 Blackwell costs $8,490 at Amazon and the AMD Radeon RX 7900 XT $800 at Amazon: the AMD Radeon RX 7900 XT is $7,690 cheaper.

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

NVIDIA RTX PRO 6000 Blackwell

$8,490

Amazon

Add to build
AMD Radeon RX 7900 XT

$800

Amazon

Add to build

Specifications

SpecNVIDIA RTX PRO 6000 BlackwellAMD Radeon RX 7900 XT
GPURTX PRO 6000 BlackwellRadeon RX 7900 XT
Memory96 GB GDDR720 GB GDDR6
Memory bus512-bit320-bit
Memory bandwidth1,792 GB/s800 GB/s
FP16 compute504 TFLOPS103 TFLOPS
SoftwareCUDAROCm
Board power600 W315 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 6000 BlackwellAMD Radeon RX 7900 XT
DeepSeek V4 Flash89tok/sUD-Q2_K_XLwon't run
Qwen 3.8 Flash Next21tok/sUD-Q3_K_XLwon't run
MiniMax M2.7117tok/sUD-Q2_K_XLwon't run
Qwen 3.8 27B37tok/sQ8_024tok/sQ4_K_M
Qwen 3.5 122B-A10B79tok/sQ6_Koffload
Qwen 3.6 27B38tok/sQ8_025tok/sQ4_K_M
gpt-oss 120B195tok/sMXFP4won't run
Qwen3 Coder Next151tok/sQ8_0won't run
Nemotron 3 Super 120B-A12B69tok/sQ5_K_Mwon't run
Gemma 4 31B33tok/sQ8_0offload
Mistral Small 4 119B110tok/sUD-Q6_Koffload
Qwen 3.6 35B-A3B201tok/sQ8_0offload
Qwen 3.5 27B38tok/sQ8_025tok/sQ4_K_M
Gemma 4 26B-A4B147tok/sQ8_078tok/sQ5_K_M
GLM 4.5 Air48tok/sQ6_Koffload
Qwen 3.5 35B-A3B201tok/sQ8_0offload
Qwen3 Next 80B-A3B196tok/sQ8_0won't run
GLM 4.7 Flash117tok/sQ8_076tok/sQ4_K_M
Devstral Small 2 24B42tok/sQ8_024tok/sQ5_K_M
Nemotron 3.5 Lightning 30B-A3B195tok/sUD-Q6_K_XLoffload
gpt-oss 20B253tok/sMXFP4100tok/sMXFP4
Kimi Linear 48B-A3B202tok/sQ8_0offload
Qwen3 Coder 30B-A3B167tok/sQ8_0100tok/sQ4_K_M
Qwen3 32B30tok/sQ8_0offload
Gemma 4 12B85tok/sQ8_034tok/sQ8_0
Qwen 3.5 9B113tok/sQ8_045tok/sQ8_0
Mistral Small 3.2 24B42tok/sQ8_024tok/sQ5_K_M
Gemma 3 27B37tok/sQ8_025tok/sQ4_K_M
DeepSeek R1 Distill Qwen 32B30tok/sQ8_0offload
Llama 3.3 70B14tok/sQ8_0offload
Ministral 3 14B71tok/sQ8_028tok/sQ8_0
Qwen3 14B65tok/sQ8_026tok/sQ8_0
DeepSeek R1 0528 Qwen3 8B112tok/sQ8_044tok/sQ8_0
Qwen3 8B112tok/sQ8_044tok/sQ8_0
Phi-4 14B64tok/sQ8_026tok/sQ8_0
Qwen 3.5 4B231tok/sQ8_092tok/sQ8_0
Ministral 3 8B109tok/sQ8_043tok/sQ8_0
Granite 4.2 8B104tok/sQ8_041tok/sQ8_0
Gemma 4 E4B131tok/sQ8_052tok/sQ8_0
LFM2.5 8B-A1B455tok/sQ8_0180tok/sQ8_0
Llama 3.1 8B145tok/sQ6_K57tok/sQ6_K
Qwen2.5 0.5B1759tok/sQ8_0697tok/sQ8_0
Qwen3.5 0.8B1193tok/sQ8_0472tok/sQ8_0
Qwen3.5 2B715tok/sQ5_K_M283tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B522tok/sQ8_0207tok/sQ8_0
MiniCPM5 2B366tok/sQ8_0145tok/sQ8_0
LFM2.5 2.6B329tok/sQ8_0130tok/sQ8_0
LFM2.5 VL 3B329tok/sQ8_0130tok/sQ8_0
Spark X2.5 4B327tok/smlx-4bit130tok/smlx-4bit
granite vision 4.1 4b259tok/sQ8_0103tok/sQ8_0
granite 4.2 3b243tok/sQ8_096tok/sQ8_0
LFM2.5 24B A2B238tok/sQ8_0119tok/sQ5_K_M
Mellum2.1 12B A2.5B Thinking236tok/sQ8_094tok/sQ8_0
FrogNano 4B 2609225tok/sQ8_089tok/sQ8_0
Kimi VL A3B Thinking 2506199tok/sQ8_079tok/sQ8_0
Kimi VL A3B190tok/smlx-8bit75tok/smlx-8bit
DeepSeek R1 Distill Qwen 7B130tok/sQ8_051tok/sQ8_0
GLM 4 9B 0414129tok/smlx-5bit51tok/smlx-5bit
DeepSeek R1 Distill Llama 8B116tok/sQ8_046tok/sQ8_0
Ministral 3 8B Reasoning 2512109tok/sQ8_043tok/sQ8_0
GLM 4.6V Flash108tok/sQ8_043tok/sQ8_0
GLM 4.1V 9B Thinking92tok/smlx-8bit36tok/smlx-8bit
GLM 4.5V67tok/sQ4_K_Mwon't run
DeepSeek R1 Distill Qwen 14B64tok/sQ8_025tok/sQ8_0
GLM 4.6V61tok/sQ4_K_Mwon't run
Qwen3.8 27B OBLITERATED52tok/smlx-4bitoffload
Devstral Small 250742tok/sQ8_024tok/sQ5_K_M
Magistral Small 250642tok/sQ8_024tok/sQ5_K_M
Magistral Small 250742tok/sQ8_024tok/sQ5_K_M
Magistral Small 250942tok/sQ8_024tok/sQ5_K_M
granite 4.2 30b33tok/sQ8_0offload
DeepSeek R1 Distill Llama 70B14tok/sQ8_0offload
Kimi Dev 72B14tok/sQ8_0won't run
Devstral 2 123B Instruct 251212tok/sQ5_K_Mwon't run
Mistral Medium 3.5 128B12tok/sQ5_K_Mwon't run
GLM 5.3 Flashoffloadwon't run
MiniMax M2.5offloadwon'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
GLM 4.5won't runwon't run
GLM 4.6won'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 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 M3won't runwon't run

Questions

Which is better for local AI, the NVIDIA RTX PRO 6000 Blackwell or the AMD Radeon RX 7900 XT?
Choose the NVIDIA RTX PRO 6000 Blackwell to run bigger models. Choose the NVIDIA RTX PRO 6000 Blackwell for speed. Choose the AMD Radeon RX 7900 XT to spend less.
Which can run bigger models, the NVIDIA RTX PRO 6000 Blackwell or the AMD Radeon RX 7900 XT?
At 8k context the NVIDIA RTX PRO 6000 Blackwell runs 75 of the models we track fully in memory and the AMD Radeon RX 7900 XT 49; only the NVIDIA RTX PRO 6000 Blackwell holds DeepSeek V4 Flash.
Which is faster for local AI, the NVIDIA RTX PRO 6000 Blackwell or the AMD Radeon RX 7900 XT?
On the models both run fully in memory, the NVIDIA RTX PRO 6000 Blackwell is about 2.5× as fast as the AMD Radeon RX 7900 XT (estimated, median across models).
What is the price difference between the NVIDIA RTX PRO 6000 Blackwell and the AMD Radeon RX 7900 XT?
In the US the NVIDIA RTX PRO 6000 Blackwell costs $8,490 at Amazon and the AMD Radeon RX 7900 XT $800 at Amazon: the AMD Radeon RX 7900 XT is $7,690 cheaper.

More comparisons