readygpu

NVIDIA GeForce RTX 4090 vs AMD Radeon RX 7900 XTX

For running models locally, priced in India. At 64k context the first runs 22 models fully in memory, the second 22.

In short

Both have 24 GB of memory for models; the NVIDIA GeForce RTX 4090 has 1,008 GB/s of bandwidth and the AMD Radeon RX 7900 XTX 960 GB/s. At 64k context both run the same 22 models we track fully in memory. On the models both run fully in memory, the NVIDIA GeForce RTX 4090 is about 1.2× as fast as the AMD Radeon RX 7900 XTX (estimated, median across models).

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

NVIDIA GeForce RTX 4090

Not priced here

Add to build
AMD Radeon RX 7900 XTX

Not priced here

Add to build

Specifications

SpecNVIDIA GeForce RTX 4090AMD Radeon RX 7900 XTX
GPUGeForce RTX 4090Radeon RX 7900 XTX
Memory24 GB GDDR6X24 GB GDDR6
Memory bus384-bit384-bit
Memory bandwidth1,008 GB/s960 GB/s
FP16 compute165 TFLOPS123 TFLOPS
SoftwareCUDAROCm
Board power450 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 GeForce RTX 4090AMD Radeon RX 7900 XTX
Qwen 3.8 27B30tok/sUD-Q4_K_M25tok/sUD-Q4_K_M
Qwen 3.6 27B26tok/sQ5_K_M25tok/sQ4_K_M
Qwen 3.6 35B-A3B115tok/sUD-Q4_K_M97tok/sUD-Q4_K_M
Qwen 3.5 27B30tok/sQ4_K_M25tok/sQ4_K_M
Gemma 4 26B-A4B78tok/sUD-Q5_K_M66tok/sUD-Q5_K_M
Qwen 3.5 35B-A3B116tok/sQ4_K_M98tok/sQ4_K_M
GLM 4.7 Flash57tok/sQ4_K_M48tok/sQ4_K_M
gpt-oss 20B93tok/sMXFP478tok/sMXFP4
Gemma 4 12B44tok/sQ8_038tok/sQ8_0
Qwen 3.5 9B53tok/sQ8_045tok/sQ8_0
Gemma 3 27B28tok/sQ4_K_M24tok/sQ4_K_M
Ministral 3 14B29tok/sQ6_K24tok/sQ6_K
Qwen3 14B27tok/sQ6_K23tok/sQ6_K
DeepSeek R1 0528 Qwen3 8B34tok/sQ8_029tok/sQ8_0
Qwen3 8B34tok/sQ8_029tok/sQ8_0
Phi-4 14B26tok/sQ5_K_M24tok/sQ4_K_M
Qwen 3.5 4B94tok/sQ8_079tok/sQ8_0
Ministral 3 8B34tok/sQ8_029tok/sQ8_0
Granite 4.2 8B31tok/sQ8_026tok/sQ8_0
Gemma 4 E4B62tok/sQ8_052tok/sQ8_0
LFM2.5 8B-A1B88tok/sQ8_074tok/sQ8_0
Llama 3.1 8B41tok/sQ6_K35tok/sQ6_K
Qwen 3.5 122B-A10Boffloadoffload
Gemma 4 31Boffloadoffload
Qwen3 Coder Nextoffloadoffload
Mistral Small 4 119Boffloadoffload
Devstral Small 2 24Boffloadoffload
Qwen3 Next 80B-A3Boffloadoffload
Nemotron 3.5 Lightning 30B-A3Boffloadoffload
Kimi Linear 48B-A3Boffloadoffload
Qwen3 Coder 30B-A3Boffloadoffload
Qwen3 32Boffloadoffload
Mistral Small 3.2 24Boffloadoffload
DeepSeek R1 Distill Qwen 32Boffloadoffload
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
GLM 4.5 Airwon't runwon't run
Llama 3.3 70Bwon't runwon't run

Questions

Which is better for local AI, the NVIDIA GeForce RTX 4090 or the AMD Radeon RX 7900 XTX?
Choose the NVIDIA GeForce RTX 4090 for speed.
Which can run bigger models, the NVIDIA GeForce RTX 4090 or the AMD Radeon RX 7900 XTX?
At 64k context both run the same 22 models we track fully in memory.
Which is faster for local AI, the NVIDIA GeForce RTX 4090 or the AMD Radeon RX 7900 XTX?
On the models both run fully in memory, the NVIDIA GeForce RTX 4090 is about 1.2× as fast as the AMD Radeon RX 7900 XTX (estimated, median across models).

More comparisons