From c530e1688076d6cd5475b4c68894e6e5b7af70e5 Mon Sep 17 00:00:00 2001
From: Frank <frank@dankware>
Date: Sat, 8 Aug 2026 17:28:00 -0700
Subject: [PATCH] Temporarily disable speculative decoding to test base model

---
 vllm-awq-qwen36-dflash-ampere.sh | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/vllm-awq-qwen36-dflash-ampere.sh b/vllm-awq-qwen36-dflash-ampere.sh
index 0d42b9b..d7a1f72 100755
--- a/vllm-awq-qwen36-dflash-ampere.sh
+++ b/vllm-awq-qwen36-dflash-ampere.sh
@@ -22,7 +22,7 @@ exec docker run --rm --name vllm-awq-qwen36-dflash-ampere \
     -v ~/.cache/huggingface:/root/.cache/huggingface \
     vllm/vllm-openai:nightly "${MODEL}" \
     --served-model-name "frank-default-small" \
-    --speculative-config '{"method": "dflash", "model": "z-lab/Qwen3.6-35B-A3B-DFlash", "num_speculative_tokens": 4}' \
+    # --speculative-config '{"method": "dflash", "model": "z-lab/Qwen3.6-35B-A3B-DFlash", "num_speculative_tokens": 4}' \
     --max-num-seqs 2 \
     --max-model-len 32768 \
     --tensor-parallel-size 1 \