List your company

Throughput Measurement on System Throughput Evaluation (test)

6,898.53Throughput (tokens/GPU/sec)GKA
1,268.70842,730.29674,191.8855,653.4733Nov 26, 2025

Evaluation results

MethodLinks
GKAScale=8B, Context Length=16K, Batch size=4M tokens2025.116,898.53
GKAScale=8B, Context Length=32K, Batch size=4M tokens2025.116,808.94
GKAScale=8B, Context Length=64K, Batch size=4M tokens2025.116,721.64
TransformerScale=8B, Context Length=16K, Batch size=4M tokens2025.116,096.37
TransformerScale=8B, Context Length=32K, Batch size=4M tokens2025.114,766.25
TransformerScale=8B, Context Length=64K, Batch size=4M tokens2025.113,382.5
GKAScale=14B, Context Length=16K, Batch size=4M tokens2025.113,318.28
GKAScale=14B, Context Length=32K, Batch size=4M tokens2025.113,256.45
GKAScale=14B, Context Length=64K, Batch size=4M tokens2025.113,256.45
TransformerScale=14B, Context Length=16K, Batch size=4M tokens2025.112,833.99
TransformerScale=14B, Context Length=32K, Batch size=4M tokens2025.112,221.56
TransformerScale=14B, Context Length=64K, Batch size=4M tokens2025.111,485.24