diff --git a/scripts/bench-spec-stack.sh b/scripts/bench-spec-stack.sh index 1f00c48..0f7893d 100644 --- a/scripts/bench-spec-stack.sh +++ b/scripts/bench-spec-stack.sh @@ -19,6 +19,11 @@ OUT_TOKENS="${3:-256}" ENV_FILE=/home/dank/vllmScripts/scripts/frank-default-large.env source "$ENV_FILE" +# Derive the PromptForge sidecar paths from the env base values. +export PROMPTFORGE_SIDECAR="$MODEL_DIR/$FFN_PFS" +export PROMPTFORGE_GDN_SIDECAR="$MODEL_DIR/$GDN_PFS" +export PROMPTFORGE_MTP_OUTPUT_K8_PROXY="$MODEL_DIR/$OUTPUT_K8_PFS" + MODEL="$MODEL_DIR/$MODEL_FILE" LLAMA_BIN=/home/dank/ciru-qwen38/ROCmFPX/build-promptforge/bin/llama-server PORT=7893 @@ -33,7 +38,7 @@ for f in "$MODEL" "$PROMPTFORGE_SIDECAR" "$PROMPTFORGE_GDN_SIDECAR" "$PROMPTFORG done [[ -x "$LLAMA_BIN" ]] || { echo "missing: $LLAMA_BIN" >&2; exit 1; } -export PROMPTFORGE_SIDECAR PROMPTFORGE_GDN_SIDECAR PROMPTFORGE_MODE +export PROMPTFORGE_MODE="m2048_fused_tail1476" export GGML_CUDA_GRAPH_OPT=0 export PROMPTFORGE_ENABLE_SMALLM_W8=1 export PROMPTFORGE_MTP_OUTPUT_K8=1