From 9617bbe814de8122b3891072b14cc87eb9983cae Mon Sep 17 00:00:00 2001 From: rock Date: Thu, 5 Nov 2026 09:41:28 +0800 Subject: [PATCH] hold clef-candle at replicas 0 until the image carries the GDN fix candle-vllm PR #5 (the conv fix) could not be merged, so CI has not published a :sm70 image containing it. Deploying clef against an older :sm70 fails silently: the delta rule returns zero on all 24 linear-attention layers and /v1/systemone still answers 200 with well-formed but wrong classifications. Raise to 1 once a :sm70 build from the fix merge exists. --- k8s/apps/llm-serving/clef-candle.yaml | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) diff --git a/k8s/apps/llm-serving/clef-candle.yaml b/k8s/apps/llm-serving/clef-candle.yaml index 248bccb..44e42ae 100644 --- a/k8s/apps/llm-serving/clef-candle.yaml +++ b/k8s/apps/llm-serving/clef-candle.yaml @@ -7,7 +7,12 @@ metadata: app.kubernetes.io/part-of: llm-serving app: clef-candle spec: - replicas: 1 + # HOLD AT 0 until candle-vllm:sm70 contains the GDN conv fix (candle-vllm PR #5). + # An older :sm70 zeroes the key channel band, starves the delta rule and makes + # every linear-attention layer return zero. The response is still HTTP 200 and + # well formed, so the failure is silent. Raise to 1 only after CI publishes a + # :sm70 build from the merge that carries the fix. + replicas: 0 strategy: # One replica holds one GPU, so Recreate keeps the rollout to a single # weights load instead of needing a second card for a surge pod.