AMD EPYC Turin 128 Core Comparison: EPYC 9745 "Zen 5C" vs. EPYC 9755 "Zen 5"

Written by Michael Larabel in Processors on 2 March 2026 at 04:41 PM EST. Page 3 of 9. 14 Comments.

👁 OpenVINO benchmark with settings of Model: Handwritten English Recognition FP16-INT8, Device: CPU, Benchmark Mode: Throughput. EPYC 9755 was the fastest.

The EPYC 9745 power efficiency story played out well too with OpenVINO and other CPU-based AI workloads. Getting much of the performance of the EPYC 9755 but at much lower power consumption via the EPYC 9745 processor.

OpenVino GenAI

👁 OpenVINO GenAI benchmark with settings of Model: TinyLlama-1.1B-Chat-v1.0, Device: CPU. EPYC 9755 was the fastest.

👁 OpenVINO GenAI benchmark with settings of Model: granite-3.0-8b-instruct, Device: CPU. EPYC 9755 was the fastest.

👁 Llama.cpp benchmark with settings of Backend: CPU BLAS, Model: granite-3.0-3b-a800m-instruct-Q8_0, Test: Prompt Processing 512. EPYC 9755 was the fastest.

👁 Llama.cpp benchmark with settings of Backend: CPU BLAS, Model: granite-3.0-3b-a800m-instruct-Q8_0, Test: Prompt Processing 2048. EPYC 9755 was the fastest.

👁 Llama.cpp benchmark with settings of Backend: CPU BLAS, Model: MiniMax-M2.5-UD-TQ1_0, Test: Prompt Processing 512. EPYC 9755 was the fastest.

👁 Llama.cpp benchmark with settings of Backend: CPU BLAS, Model: MiniMax-M2.5-UD-TQ1_0, Test: Prompt Processing 2048. EPYC 9755 was the fastest.

For many of the AI workloads, the 320 Watt cTDP didn't end up resulting in much of a power/performance difference.

👁 vLLM CPU benchmark with settings of Test: Hermes-3-Llama-3.2-3B Throughput. EPYC 9755 was the fastest.

👁 vLLM CPU benchmark with settings of Test: Hermes-3-Llama-3.2-3B Latency. EPYC 9755 was the fastest.

👁 vLLM CPU benchmark with settings of Test: Hermes-3-Llama-3.1-8B Throughput. EPYC 9755 was the fastest.

👁 vLLM CPU benchmark with settings of Test: Hermes-3-Llama-3.1-8B Latency. EPYC 9755 was the fastest.

👁 vLLM CPU benchmark with settings of Test: Qwen2-VL-7B-Instruct Latency. EPYC 9755 was the fastest.

👁 vLLM CPU benchmark with settings of Test: deepseek-moe-16b-chat Latency. EPYC 9755 was the fastest.

For vLLM the latency was similar whether using the EPYC 9755 or EPYC 9745.

14 Comments - Next Page

Page: 1 2 3 4 5 6 7 8 9 Next Page

URL: https://www.phoronix.com/review/amd-epyc-9745-9755/3

⇱ AMD EPYC Turin 128 Core Comparison: EPYC 9745 "Zen 5C" vs. EPYC 9755 "Zen 5" Review - Phoronix

AMD EPYC Turin 128 Core Comparison: EPYC 9745 "Zen 5C" vs. EPYC 9755 "Zen 5"