Home › Evidence › Records › cfg-0015

cfg-0015

backfilled
citable URL: https://halobench.com/records/cfg-0015/ — this address never moves; the anchor /records/#cfg-0015 keeps resolving
modelNVIDIA-Nemotron-3-Super-120B-A12B-UD-IQ4_XS-00001-of-00003.gguf · UD-IQ4_XS
engineggml-org/llama.cpp 3653e6d · rocm · host aihydra (igpu)
flags-ngl 999 -fa 1 -b 2048 -ub 512 -ctk f16 -ctv f16 -t 16
templatenot recorded at test time
treeupstream — stock
aged evidence — reconstructed from the archive. Ingested from llama-bench, which measures throughput and not footprint: the memory block, chat-template hash and model revision are absent by construction. Peak memory comes from a Tier 0 capability probe, not from this tool. Everything present here — build commit, backend, KV quant, batch sizes, flash attention — is machine-reported.

Capability basis

unmeasured — performance numbers on this config stand on a guard alone, not an established capability

Runs on this config (6)

rundatekindsuitemetrics
run-00382026-08-08performancellama-bench@1prefill_tps=225.356478
run-00392026-08-08performancellama-bench@1decode_tps=17.505975
run-00402026-08-08performancellama-bench@1prefill_tps=226.937142
run-00412026-08-08performancellama-bench@1decode_tps=17.479771
run-00422026-08-08performancellama-bench@1prefill_tps=212.182249
run-00432026-08-08performancellama-bench@1decode_tps=17.16415

Cited by — computed at build time, never stored

model pages nemotron3-super