cfg-0113
backfilled
citable URL: https://halobench.com/records/cfg-0113/ — this address never moves; the anchor /records/#cfg-0113 keeps resolving
| model | Qwen3-Coder-Next-Q8_0-00001-of-00004.gguf · Q8_0 |
| engine | ggml-org/llama.cpp 3653e6d6d · vulkan · host aihydra (igpu) |
| flags | -ngl 999 -fa 1 -b 2048 -ub 512 -ctk f16 -ctv f16 -t 16 |
| template | not recorded at test time |
| tree | upstream — 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 (18)
| run | date | kind | suite | metrics |
|---|---|---|---|---|
| run-0414 | 2026-08-17 | performance | llama-bench@1 | prefill_tps=613.999119 |
| run-0415 | 2026-08-17 | performance | llama-bench@1 | decode_tps=44.133323 |
| run-0416 | 2026-08-17 | performance | llama-bench@1 | prefill_tps=616.105913 |
| run-0417 | 2026-08-17 | performance | llama-bench@1 | decode_tps=44.085037 |
| run-0418 | 2026-08-17 | performance | llama-bench@1 | prefill_tps=617.627368 |
| run-0419 | 2026-08-17 | performance | llama-bench@1 | decode_tps=44.193895 |
| run-0420 | 2026-08-17 | performance | llama-bench@1 | prefill_tps=204.973926 |
| run-0421 | 2026-08-17 | performance | llama-bench@1 | decode_tps=26.402386 |
| run-0422 | 2026-08-17 | performance | llama-bench@1 | prefill_tps=98.753953 |
| run-0423 | 2026-08-17 | performance | llama-bench@1 | decode_tps=21.725077 |
| run-0424 | 2026-08-17 | performance | llama-bench@1 | prefill_tps=68.766117 |
| run-0425 | 2026-08-17 | performance | llama-bench@1 | decode_tps=19.15177 |
| run-0426 | 2026-08-17 | performance | llama-bench@1 | prefill_tps=450.494255 |
| run-0427 | 2026-08-17 | performance | llama-bench@1 | decode_tps=36.971459 |
| run-0428 | 2026-08-17 | performance | llama-bench@1 | prefill_tps=450.301264 |
| run-0429 | 2026-08-17 | performance | llama-bench@1 | decode_tps=37.062648 |
| run-0430 | 2026-08-17 | performance | llama-bench@1 | prefill_tps=447.507683 |
| run-0431 | 2026-08-17 | performance | llama-bench@1 | decode_tps=36.971757 |
Cited by — computed at build time, never stored
model pages qwen3-coder-next-80b