readygpu

NVIDIA RTX A6000 vs AMD Radeon RX 7900 XTX

For running models locally, priced in the US. At 128k context the first runs 61 models fully in memory, the second 30.

In short

The NVIDIA RTX A6000 has more memory for models: 48 GB against 24 GB. At 128k context the NVIDIA RTX A6000 runs 61 of the models we track fully in memory and the AMD Radeon RX 7900 XTX 30; only the NVIDIA RTX A6000 holds Qwen 3.8 27B. On the models both run fully in memory, the AMD Radeon RX 7900 XTX is about 1.1× as fast as the NVIDIA RTX A6000 (estimated, median across models). In the US the NVIDIA RTX A6000 costs $5,989 at Amazon and the AMD Radeon RX 7900 XTX $1,300 at Amazon: the AMD Radeon RX 7900 XTX is $4,689 cheaper.

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

NVIDIA RTX A6000

$5,989

Amazon

Add to build
AMD Radeon RX 7900 XTX

$1,300

Amazon

Add to build

Specifications

SpecNVIDIA RTX A6000AMD Radeon RX 7900 XTX
GPURTX A6000Radeon RX 7900 XTX
Memory48 GB GDDR624 GB GDDR6
Memory bus384-bit384-bit
Memory bandwidth768 GB/s960 GB/s
FP16 compute38.71 TFLOPS123 TFLOPS
SoftwareCUDAROCm
Board power300 W355 W
PCIe generationGen 4Gen 4
PCIe lanesx16x16

What each runs

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

ModelNVIDIA RTX A6000AMD Radeon RX 7900 XTX
Qwen 3.8 27B13tok/sQ8_0offload
Qwen 3.5 122B-A10B43tok/sUD-Q2_K_XLoffload
Qwen 3.6 27B13tok/sQ8_0offload
Gemma 4 31B11tok/sQ8_0offload
Mistral Small 4 119B53tok/sUD-Q2_K_XLoffload
Qwen 3.6 35B-A3B49tok/sQ8_0offload
Qwen 3.5 27B13tok/sQ8_0offload
Gemma 4 26B-A4B41tok/sQ8_054tok/sQ5_K_M
Qwen 3.5 35B-A3B49tok/sQ8_0offload
GLM 4.7 Flash23tok/sQ8_0offload
Devstral Small 2 24B10tok/sQ8_0offload
Nemotron 3.5 Lightning 30B-A3B68tok/sUD-Q6_K_XLoffload
Kimi Linear 48B-A3B80tok/sQ6_Koffload
gpt-oss 20B51tok/sMXFP456tok/sMXFP4
Qwen3 Coder 30B-A3B18tok/sQ8_0offload
Gemma 4 12B31tok/sQ8_035tok/sQ8_0
Qwen 3.5 9B34tok/sQ8_038tok/sQ8_0
Gemma 3 27B12tok/sQ8_0offload
Mistral Small 3.2 24B10tok/sQ8_0offload
Ministral 3 14B13tok/sQ8_0offload
Qwen3 14B13tok/sQ8_0offload
DeepSeek R1 0528 Qwen3 8B17tok/sQ8_0offload
Qwen3 8B17tok/sQ8_0offload
Phi-4 14B11tok/sQ8_0offload
Qwen 3.5 4B54tok/sQ8_060tok/sQ8_0
Ministral 3 8B18tok/sQ8_0offload
Granite 4.2 8B15tok/sQ8_0offload
Gemma 4 E4B40tok/sQ8_044tok/sQ8_0
LFM2.5 8B-A1B38tok/sQ8_042tok/sQ8_0
Llama 3.1 8B20tok/sQ6_K23tok/sQ5_K_M
Qwen2.5 0.5B222tok/sQ8_0247tok/sQ8_0
Qwen3.5 0.8B195tok/sQ8_0216tok/sQ8_0
Qwen3.5 2B155tok/sQ5_K_M172tok/sQ5_K_M
LFM3 1.7B137tok/sQ8_0152tok/sQ8_0
DeepSeek R1 Distill Qwen 1.5B84tok/sQ8_093tok/sQ8_0
Mellum2.1 12B A2.5B Thinking62tok/sQ8_069tok/sQ8_0
Spark X2.5 4B60tok/smlx-4bit67tok/smlx-4bit
MiniCPM5 2B57tok/sQ8_064tok/sQ8_0
FrogNano 4B 260953tok/sQ8_059tok/sQ8_0
LFM2.5 2.6B44tok/sQ8_048tok/sQ8_0
LFM2.5 VL 3B44tok/sQ8_048tok/sQ8_0
Kimi VL A3B Thinking 250640tok/sQ8_044tok/sQ8_0
Kimi VL A3B39tok/smlx-8bit43tok/smlx-8bit
GLM 4 9B 041435tok/smlx-5bit39tok/smlx-5bit
granite vision 4.1 4b33tok/sQ8_037tok/sQ8_0
granite 4.2 3b33tok/sQ8_036tok/sQ8_0
GLM 4.6V Flash31tok/sQ8_034tok/sQ8_0
DeepSeek R1 Distill Qwen 7B31tok/sQ8_034tok/sQ8_0
GLM 4.1V 9B Thinking28tok/smlx-8bit31tok/smlx-8bit
Ministral 3 3B Instruct 251227tok/sQ8_030tok/sQ8_0
LFM2.5 24B A2B22tok/sQ8_0offload
DeepSeek R1 Distill Llama 8B19tok/sQ8_023tok/sQ5_K_M
Ministral 3 8B Reasoning 251218tok/sQ8_0offload
Qwen3.8 27B OBLITERATED16tok/smlx-4bitoffload
Muse Glimmer 30B14tok/sQ8_024tok/sQ5_K_M
Ministral 3 14B Reasoning 251213tok/sQ8_0offload
DeepSeek R1 Distill Qwen 14B12tok/sQ8_0offload
Devstral Small 250710tok/sQ8_0offload
Magistral Small 250610tok/sQ8_0offload
Magistral Small 250710tok/sQ8_0offload
Magistral Small 250910tok/sQ8_0offload
Qwen3 Coder Nextoffloadwon't run
Nemotron 3 Super 120B-A12Boffloadwon't run
gpt-oss 120Boffloadwon't run
GLM 4.5 Airoffloadwon't run
Qwen3 Next 80B-A3Boffloadwon't run
Qwen3 32Boffloadwon't run
DeepSeek R1 Distill Qwen 32Boffloadwon't run
granite 4.2 30boffloadwon'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
Llama 3.3 70Bwon't runwon't run
DeepSeek R1won't runwon't run
DeepSeek R1 0528won't runwon't run
DeepSeek R1 Distill Llama 70Bwon'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 A6000 or the AMD Radeon RX 7900 XTX?
Choose the NVIDIA RTX A6000 to run bigger models. Choose the AMD Radeon RX 7900 XTX for speed. Choose the AMD Radeon RX 7900 XTX to spend less.
Which can run bigger models, the NVIDIA RTX A6000 or the AMD Radeon RX 7900 XTX?
At 128k context the NVIDIA RTX A6000 runs 61 of the models we track fully in memory and the AMD Radeon RX 7900 XTX 30; only the NVIDIA RTX A6000 holds Qwen 3.8 27B.
Which is faster for local AI, the NVIDIA RTX A6000 or the AMD Radeon RX 7900 XTX?
On the models both run fully in memory, the AMD Radeon RX 7900 XTX is about 1.1× as fast as the NVIDIA RTX A6000 (estimated, median across models).
What is the price difference between the NVIDIA RTX A6000 and the AMD Radeon RX 7900 XTX?
In the US the NVIDIA RTX A6000 costs $5,989 at Amazon and the AMD Radeon RX 7900 XTX $1,300 at Amazon: the AMD Radeon RX 7900 XTX is $4,689 cheaper.

More comparisons

More hardware for local AI