inference/research
verifiedmedium · 0.75updated 2026-10-15

Argonne MI300X llama.cpp Qwen2-7B b1

70.3 tok/s aggregate (batch*(in+out)/lat)

Fields

id
benchmarks/argonne-mi300x-llamacpp-qwen2-7b-fp16-b1-128io-tps
Metric
tps_aggregate
Value
70.3
Unit
tok/s aggregate (batch*(in+out)/lat)
Model
Qwen/Qwen2-7B
Methodology
THIRD-PARTY MEASURED (Argonne LLM-Inference-Bench arXiv:2411.00136, All_results.csv). llama.cpp, FP16/BF16, 128in/128out, AGGREGATE over batch. 1x bs1 lat 3.64s. CSV 'AMD MI300X GPU' = 8x MI300X node (xGMI Infinity Fabric).
Measured by
third_party
Reproducible
yes

Cross-references