frank-default-small: enable prefix caching (opt-in for hybrid models; big TTFT win for pi's repeated system prompt)

scripts/frank-default-small.env

@@ -39,6 +39,14 @@ MAX_NUM_BATCHED_TOKENS=8192
 # keeps headroom for the host.
 GPU_MEMORY_UTILIZATION=0.92
 
+# Prefix caching. This model is hybrid (GatedDeltaNet + full attention); vLLM
+# keeps prefix caching OPT-IN by default for hybrid models ("while the feature
+# matures"), so it is off unless we ask for it. Turning it on reuses the KV of
+# the shared prompt prefix (pi re-sends the same system prompt + tool schema
+# every turn), so follow-up requests skip re-prefilling thousands of tokens.
+# Big TTFT win for the agent loop. Set 0 to disable.
+ENABLE_PREFIX_CACHING=1
+
 # ngram speculative decoding (vLLM ngram proposer).
 # prompt_lookup_max = ngram window; num_speculative_tokens = draft depth.
 NGRAM_LOOKUP_MAX=5