Decode · tok/s
127
Prefill
—
tok/100W
52.9
GLM 5.3 Flash on 4 DGX Sparks with TensorFold vs my vLLM TP4 stack, same harness: Code 127 vs 120 tok/s JSON 133 vs 120 Tool calls 93 vs 68 (+37%) Faster at every concurrency, C1 to C6 500K-token prompt in 191s (vLLM 202s) 8 streams, 500K context each, vision on
First-person 4× Spark TensorFold vs own vLLM. Rank is the quoted 127 code. JSON and tool numbers stay in the note. Quant not named.
