cfg-0141
backfilled
citable URL: https://halobench.com/records/cfg-0141/ — this address never moves; the anchor /records/#cfg-0141 keeps resolving
| model | Qwen3.6-35B-A3B-MTP-UD-Q4_K_M.gguf · UD-Q4_K_M · rev 0b21525e972670ed59e1812e170b27c26355381f0656ecc4e25617ece7dac58b |
| engine | ggml-org/llama.cpp 7077abbe14c510cb829c93a1328c2815b5805ebd · ROCm0/gfx1151 · host aihydra (igpu) |
| flags | HU-004 long-prompt guard plain/control arm on raw llama-server: model /home/aihydra/models/qwen3.6-35b-a3b-mtp/Qwen3.6-35B-A3B-UD-Q4_K_M.gguf, -ngl 999 -dev ROCm0 -fa on -ctk f16 -ctv f16 -c 204800 --load-mode none -b 2048 -ub 512 --parallel 1 --host 127.0.0.1 --port 8114 --spec-type none. Server binary build-rocm/bin/llama-server, build 50 commit 7077abb — the same binary used by HO-009 cfg-0132/cfg-0134. No proxy/llama-swap/ngram, no MTP/NextN on this plain control arm. |
| template | not recorded at test time |
| tree | upstream — stock |
aged evidence — reconstructed from the archive. HU-004 reviewer-admitted long-prompt guard control arm only. Guard-cleared evidence (all long-prompt fixtures non-empty content / non-empty required tool arguments, no silent empty success) is a guard-pass result, not a capability, throughput, or production result. This config does not establish agentic/tau2 capability or served-path production performance.
Capability basis
unmeasured — performance numbers on this config stand on a guard alone, not an established capability
Runs on this config (1)
| run | date | kind | suite | metrics |
|---|---|---|---|---|
| run-0551 | 2026-08-21 | guard | guard-hu004-longprompt-8kto40k | tool_call_success=1 · empty_arg_calls=0 · tasks_passed=10 · tasks_total=10 |
Cited by — computed at build time, never stored
claims clm-0101