After getting beaten by the Radeon RX 7900 XTX in DeepSeek AI benchmarks that AMD published, Nvidia has come back swinging, claiming its RTX 5090 and RTX 4090 GPUs are significantly faster than the RDNA 3 flagship. Nvidia countered in a blog post that the RTX 5090 is up to 2.2x faster than the RX 7900 XTX.
Nvidia benchmarked the RTX 5090, RTX 4090, and RX 7900 XTX in three DeepSeek R1 AI model versions, using Distill Qwen 7b, Llama 8b, and Qwen 32b. Using the Qwen LLM with the 32b parameter, the RTX 5090 was allegedly 124% faster, and the RTX 4090 47% faster than the RX 7900 XTX. Using Llama 8b, the RTX 5090 was 106% faster, and the RTX 4090 was 47% faster than the RX 7900 XTX. Using Qwen 7b, the RTX 5090 was 103% quicker, and the RTX 4090 was 46% more performant than the RX 7900 XTX.
Nvidia’s results are a slap in the face to AMD’s own benchmarks featuring the RTX 4090 and RTX 4080. The RX 7900 XTX was faster than both Ada Lovelace GPUs except for one instance, where it was a few percent slower than the RTX 4090. The RX 7900 XTX was up to 113% quicker and 134% faster than the RTX 4090 and RTX 4080, respectively, according to AMD.
Nvidia provides a significantly different picture with the RTX 4090, showing that the RTX 4090 is considerably faster than the RX 7900 XTX, not the other way around. The icing on the cake (for Nvidia) is that the RTX 5090 more than doubled the RTX 4090’s performance results, thoroughly crushing the RX 7900 XTX.
This is a good reminder to take manufacturer-provided benchmark results with a pinch of salt. Nvidia’s results might not even be correct, but regardless, AMD and Nvidia provide results that contradict each other.
DeepSeek R1 is a new AI model that has blown away the industry, providing competitive performance with the best AI models out there but requiring 11 times less computing power. DeepSeek’s incredible achievement was solely responsible for Nvidia losing nearly $600 billion in market capital in a single day.
