diff --git a/Makefile b/Makefile
index 0f10f1a..e1c5411 100644
--- a/Makefile
+++ b/Makefile
@@ -1590,13 +1590,25 @@ cloud-fetch-chunk: bootstrap ## fetch + hash-verify one chunk from a bucket [LEA
 	@test -n "$(BLOB_BASE)" || { echo 'BLOB_BASE required (per-chunk blobs/<hash> base URL)'; exit 2; }
 	@$(ARBORIST) cloud fetch-chunk "$(LEAF_HASH)" --blob-base "$(BLOB_BASE)"
 
-cloud-query: bootstrap ## bucket-direct end-to-end via manifest [Q="..." JSON=1 BUCKET_URL=... TOP_K=N MAX_CONTEXT=N CACHE_MB=N]
-	@test -n "$(Q)" || { echo 'usage: make cloud-query Q="your question" [JSON=1] [BUCKET_URL=...] [TOP_K=4] [MAX_CONTEXT=24000] [CACHE_MB=64]'; exit 2; }
+# LLM toggle: `make cloud-query LLM=qwen Q="..."` swaps to Qwen3.6-27B on
+# uncloseai. Default (LLM unset or LLM=hermes) leaves --endpoint/--model
+# unset so cloud-query falls back to its built-in default (Hermes-3-8B on
+# ai.unturf.com). Per-call LLM_ENDPOINT=… / LLM_MODEL=… still override.
+ifeq ($(LLM),qwen)
+LLM_ENDPOINT ?= https://qwen.ai.unturf.com/v1
+LLM_MODEL    ?= Qwen3.6-27B-UD-Q4_K_XL.gguf
+endif
+
+cloud-query: bootstrap ## bucket-direct end-to-end via manifest [Q="..." JSON=1 LLM=qwen|hermes BUCKET_URL=... TOP_K=N MAX_CONTEXT=N CACHE_MB=N]
+	@test -n "$(Q)" || { echo 'usage: make cloud-query Q="your question" [JSON=1] [LLM=qwen|hermes] [BUCKET_URL=...] [TOP_K=4] [MAX_CONTEXT=24000] [CACHE_MB=64]'; exit 2; }
 	@echo "# bucket: $(BUCKET_URL)" >&2
+	$(if $(LLM_ENDPOINT),@echo "# llm:    $(LLM_ENDPOINT) / $(LLM_MODEL)" >&2,)
 	@$(ARBORIST) cloud query '$(Q)' --bucket-url "$(BUCKET_URL)" \
 	    --top-k $(or $(TOP_K),4) \
 	    --max-context-chars $(or $(MAX_CONTEXT),24000) \
 	    --cache-mb $(or $(CACHE_MB),64) \
+	    $(if $(LLM_ENDPOINT),--endpoint $(LLM_ENDPOINT),) \
+	    $(if $(LLM_MODEL),--model $(LLM_MODEL),) \
 	    $(if $(JSON),--json,)
 
 # ---------------------------------------------------------------------------