cfg-0055
backfilled
citable URL: https://halobench.com/records/cfg-0055/ — this address never moves; the anchor /records/#cfg-0055 keeps resolving
| model | Qwen3.6-35B-A3B-UD-Q4_K_XL.gguf · UD-Q4_K_XL |
| engine | ggml-org/llama.cpp a94d563 · rocm · 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 (6)
| run | date | kind | suite | metrics |
|---|---|---|---|---|
| run-0222 | 2026-08-14 | performance | llama-bench@27 | prefill_tps=1063.898925 |
| run-0223 | 2026-08-14 | performance | llama-bench@27 | decode_tps=50.733048 |
| run-0224 | 2026-08-14 | performance | llama-bench@27 | prefill_tps=254.950374 |
| run-0225 | 2026-08-14 | performance | llama-bench@27 | decode_tps=28.333051 |
| run-0226 | 2026-08-14 | performance | llama-bench@27 | prefill_tps=593.669237 |
| run-0227 | 2026-08-14 | performance | llama-bench@27 | decode_tps=42.1234 |