readygpu

NVIDIA GeForce RTX 5090 vs AMD Radeon RX 7900 XTX

For running models locally, priced in Germany. At 64k context the first runs 56 models fully in memory, the second 45.

In short

The NVIDIA GeForce RTX 5090 has more memory for models: 32 GB against 24 GB. At 64k context the NVIDIA GeForce RTX 5090 runs 56 of the models we track fully in memory and the AMD Radeon RX 7900 XTX 45; only the NVIDIA GeForce RTX 5090 holds Gemma 4 31B. On the models both run fully in memory, the NVIDIA GeForce RTX 5090 is about 2.1× as fast as the AMD Radeon RX 7900 XTX (estimated, median across models). In Germany the NVIDIA GeForce RTX 5090 costs 6.444 € at Amazon and the AMD Radeon RX 7900 XTX 6.587 € at Amazon: the NVIDIA GeForce RTX 5090 is 143 € cheaper.

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

NVIDIA GeForce RTX 5090

6.444 €

Amazon

Add to build
AMD Radeon RX 7900 XTX

6.587 €

Amazon

Add to build

Specifications

SpecNVIDIA GeForce RTX 5090AMD Radeon RX 7900 XTX
GPUGeForce RTX 5090Radeon RX 7900 XTX
Memory32 GB GDDR724 GB GDDR6
Memory bus512-bit384-bit
Memory bandwidth1,792 GB/s960 GB/s
FP16 compute209 TFLOPS123 TFLOPS
SoftwareCUDAROCm
Board power575 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 GeForce RTX 5090AMD Radeon RX 7900 XTX
Qwen 3.8 27B39tok/sQ6_K24tok/sQ4_K_M
Qwen 3.6 27B41tok/sQ6_K25tok/sQ4_K_M
Gemma 4 31B35tok/sQ6_Koffload
Qwen 3.6 35B-A3B167tok/sQ6_K97tok/sUD-Q4_K_M
Qwen 3.5 27B41tok/sQ6_K25tok/sQ4_K_M
Gemma 4 26B-A4B117tok/sQ8_066tok/sUD-Q5_K_M
Qwen 3.5 35B-A3B175tok/sQ6_K98tok/sQ4_K_M
GLM 4.7 Flash87tok/sQ6_K48tok/sQ4_K_M
Devstral Small 2 24B37tok/sQ6_Koffload
Nemotron 3.5 Lightning 30B-A3B201tok/sUD-Q5_K_Moffload
Kimi Linear 48B-A3B277tok/sQ4_K_Moffload
gpt-oss 20B165tok/sMXFP478tok/sMXFP4
Qwen3 Coder 30B-A3B75tok/sQ6_Koffload
Gemma 4 12B79tok/sQ8_038tok/sQ8_0
Qwen 3.5 9B95tok/sQ8_045tok/sQ8_0
Gemma 3 27B40tok/sQ6_K24tok/sQ4_K_M
Mistral Small 3.2 24B37tok/sQ6_Koffload
Ministral 3 14B44tok/sQ8_024tok/sQ6_K
Qwen3 14B42tok/sQ8_023tok/sQ6_K
DeepSeek R1 0528 Qwen3 8B61tok/sQ8_029tok/sQ8_0
Qwen3 8B61tok/sQ8_029tok/sQ8_0
Phi-4 14B38tok/sQ8_024tok/sQ4_K_M
Qwen 3.5 4B166tok/sQ8_079tok/sQ8_0
Ministral 3 8B61tok/sQ8_029tok/sQ8_0
Granite 4.2 8B55tok/sQ8_026tok/sQ8_0
Gemma 4 E4B110tok/sQ8_052tok/sQ8_0
LFM2.5 8B-A1B156tok/sQ8_074tok/sQ8_0
Llama 3.1 8B73tok/sQ6_K35tok/sQ6_K
Qwen2.5 0.5B831tok/sQ8_0395tok/sQ8_0
Qwen3.5 0.8B679tok/sQ8_0323tok/sQ8_0
Qwen3.5 2B492tok/sQ5_K_M234tok/sQ5_K_M
DeepSeek R1 Distill Qwen 1.5B294tok/sQ8_0140tok/sQ8_0
MiniCPM5 2B202tok/sQ8_096tok/sQ8_0
Spark X2.5 4B202tok/smlx-4bit96tok/smlx-4bit
Mellum2.1 12B A2.5B Thinking183tok/sQ8_087tok/sQ8_0
FrogNano 4B 2609163tok/sQ8_077tok/sQ8_0
LFM2.5 2.6B161tok/sQ8_077tok/sQ8_0
LFM2.5 VL 3B161tok/sQ8_077tok/sQ8_0
Kimi VL A3B Thinking 2506130tok/sQ8_062tok/sQ8_0
Kimi VL A3B126tok/smlx-8bit60tok/smlx-8bit
granite vision 4.1 4b124tok/sQ8_059tok/sQ8_0
granite 4.2 3b120tok/sQ8_057tok/sQ8_0
GLM 4 9B 0414102tok/smlx-5bit48tok/smlx-5bit
DeepSeek R1 Distill Qwen 7B94tok/sQ8_045tok/sQ8_0
LFM2.5 24B A2B89tok/sQ8_048tok/sQ4_K_M
GLM 4.6V Flash88tok/sQ8_042tok/sQ8_0
GLM 4.1V 9B Thinking77tok/smlx-8bit37tok/smlx-8bit
DeepSeek R1 Distill Llama 8B65tok/sQ8_031tok/sQ8_0
Ministral 3 8B Reasoning 251261tok/sQ8_029tok/sQ8_0
Muse Glimmer 30B46tok/sQ6_K25tok/sQ5_K_M
Qwen3.8 27B OBLITERATED44tok/smlx-4bitoffload
Devstral Small 250740tok/sQ5_K_Moffload
Magistral Small 250740tok/sQ5_K_Moffload
Magistral Small 250940tok/sQ5_K_Moffload
DeepSeek R1 Distill Qwen 14B39tok/sQ8_023tok/sQ5_K_M
Magistral Small 250637tok/sQ6_Koffload
Qwen 3.5 122B-A10Boffloadoffload
Qwen3 Coder Nextoffloadoffload
Nemotron 3 Super 120B-A12Boffloadwon't run
Mistral Small 4 119Boffloadoffload
GLM 4.5 Airoffloadwon't run
Qwen3 Next 80B-A3Boffloadoffload
Qwen3 32Boffloadoffload
DeepSeek R1 Distill Qwen 32Boffloadoffload
granite 4.2 30boffloadoffload
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
gpt-oss 120Bwon'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 GeForce RTX 5090 or the AMD Radeon RX 7900 XTX?
Choose the NVIDIA GeForce RTX 5090 to run bigger models. Choose the NVIDIA GeForce RTX 5090 for speed. Choose the NVIDIA GeForce RTX 5090 to spend less.
Which can run bigger models, the NVIDIA GeForce RTX 5090 or the AMD Radeon RX 7900 XTX?
At 64k context the NVIDIA GeForce RTX 5090 runs 56 of the models we track fully in memory and the AMD Radeon RX 7900 XTX 45; only the NVIDIA GeForce RTX 5090 holds Gemma 4 31B.
Which is faster for local AI, the NVIDIA GeForce RTX 5090 or the AMD Radeon RX 7900 XTX?
On the models both run fully in memory, the NVIDIA GeForce RTX 5090 is about 2.1× as fast as the AMD Radeon RX 7900 XTX (estimated, median across models).
What is the price difference between the NVIDIA GeForce RTX 5090 and the AMD Radeon RX 7900 XTX?
In Germany the NVIDIA GeForce RTX 5090 costs 6.444 € at Amazon and the AMD Radeon RX 7900 XTX 6.587 € at Amazon: the NVIDIA GeForce RTX 5090 is 143 € cheaper.

More comparisons

More hardware for local AI