Home › Evidence › Records › cfg-0016

cfg-0016

backfilled
citable URL: https://halobench.com/records/cfg-0016/ — this address never moves; the anchor /records/#cfg-0016 keeps resolving
modelgpt-oss-120b-UD-Q4_K_XL-00001-of-00002.gguf · UD-Q4_K_XL
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-00442026-08-08performancellama-bench@1prefill_tps=450.832446
run-00452026-08-08performancellama-bench@1decode_tps=55.454588
run-00462026-08-08performancellama-bench@1prefill_tps=421.546647
run-00472026-08-08performancellama-bench@1decode_tps=52.793591
run-00482026-08-08performancellama-bench@1prefill_tps=308.546401
run-00492026-08-08performancellama-bench@1decode_tps=41.698106

Cited by — computed at build time, never stored

model pages gpt-oss-120b