[https://nvbugs/5427043][fix] cherrypick: request length exceeds max_num_tokens (NVIDIA#7718)

Signed-off-by: Superjomn <[email protected]>
Signed-off-by: Wangshanshan <[email protected]>

Author: Superjomn

tests/unittest/llmapi/test_llm_pytorch.py

@@ -892,3 +892,17 @@ def test_min_tokens(use_speculative: bool):
 
     assert len(res.outputs) == 1
     assert len(res.outputs[0].token_ids) == output_len
+
+
+class TestLlmError:
+
+    def test_max_num_token_check(self):
+        """ LLM should raise error when got prompt length exceed the valid range. """
+        llm = LLM(llama_model_path,
+                  kv_cache_config=global_kvcache_config,
+                  max_num_tokens=100)
+
+        with pytest.raises(ValueError,
+                           match="should not exceed max_num_tokens"):
+            ids = [random.randint(10, 100) for _ in range(101)]
+            llm.generate([ids])