diff --git a/c/Makefile b/c/Makefile
index de8313c..1e0b1ea 100644
--- a/c/Makefile
+++ b/c/Makefile
@@ -146,7 +146,7 @@ else
 PYTHON    ?= python3
 endif
 CUDA_OBJ  =
-TEST_BINS = tests/test_json$(EXE) tests/test_st$(EXE) tests/test_tier$(EXE) tests/test_grammar$(EXE) tests/test_schema_gbnf$(EXE) tests/test_decode_batch$(EXE) tests/test_idot$(EXE) tests/test_i4_acc512$(EXE) tests/test_compat_direct$(EXE)
+TEST_BINS = tests/test_json$(EXE) tests/test_st$(EXE) tests/test_tier$(EXE) tests/test_grammar$(EXE) tests/test_schema_gbnf$(EXE) tests/test_decode_batch$(EXE) tests/test_idot$(EXE) tests/test_kv_alloc$(EXE) tests/test_i4_acc512$(EXE) tests/test_compat_direct$(EXE)
 
 # Windows CUDA DLL path: host links the loader, NOT cudart.
 ifneq ($(IS_WIN),)
@@ -260,6 +260,9 @@ tests/test_decode_batch$(EXE): tests/test_decode_batch.c decode_batch.h
 tests/test_idot$(EXE): tests/test_idot.c glm.c st.h json.h tok.h tok_unicode.h compat.h grammar.h tier.h
 	$(CC) $(CFLAGS) $< -o $@ $(LDFLAGS)
 
+tests/test_kv_alloc$(EXE): tests/test_kv_alloc.c glm.c st.h json.h tok.h tok_unicode.h compat.h grammar.h tier.h
+	$(CC) $(CFLAGS) $< -o $@ $(LDFLAGS)
+
 tests/test_i4_acc512$(EXE): tests/test_i4_acc512.c
 	$(CC) $(CFLAGS) $< -o $@ $(LDFLAGS)
 
diff --git a/c/tests/test_kv_alloc.c b/c/tests/test_kv_alloc.c
new file mode 100644
index 0000000..40ec00d
--- /dev/null
+++ b/c/tests/test_kv_alloc.c
@@ -0,0 +1,23 @@
+/* kv_alloc must survive re-allocation on the same KVState: every free path is
+ * guarded by if(k->Lc) precisely so callers (context resize, slot re-init) can
+ * call it again. A stale duplicate free block frees every Lc[i]/Rc[i] and both
+ * arrays twice on the second call -> allocator abort. No model file needed:
+ * the CPU path of kv_alloc only reads c->n_layers/kv_lora/qk_rope. */
+#define main coli_glm_main_unused
+#include "../glm.c"
+#undef main
+
+int main(void){
+    static Model m;
+    m.c.n_layers=2; m.c.kv_lora=8; m.c.qk_rope=4;
+    m.kv=calloc(1,sizeof(KVState));
+    kv_alloc(&m,16);
+    for(int i=0;i<m.c.n_layers+1;i++){ m.Lc[i][0]=1.0f; m.Rc[i][0]=1.0f; }
+    kv_alloc(&m,32);                       /* the re-allocation path under test */
+    for(int i=0;i<m.c.n_layers+1;i++){
+        m.Lc[i][(int64_t)32*m.c.kv_lora-1]=2.0f;
+        m.Rc[i][(int64_t)32*m.c.qk_rope-1]=2.0f;
+    }
+    printf("OK kv_alloc re-allocation\n");
+    return 0;
+}