fix: passing max_length to vllm engine args (#1124)

* fix: passing max_length to vllm engine args

* feat: add `max_model_len`

* chore: lint

lm_eval/models/vllm_causallms.py

@@ -46,6 +46,7 @@ class VLLM(LM):
         batch_size: Union[str, int] = 1,
         max_batch_size=None,
         max_length: int = None,
+        max_model_len: int = None,
         seed: int = 1234,
         gpu_memory_utilization: float = 0.9,
         device: str = "cuda",
@@ -62,6 +63,11 @@ please install vllm via `pip install lm-eval[vllm]` or `pip install -e .[vllm]`"
             )
 
         assert "cuda" in device or device is None, "vLLM only supports CUDA"
+        assert (
+            max_length is None or max_model_len is None
+        ), "Either max_length or max_model_len may be provided, but not both"
+
+        self._max_length = max_model_len if max_model_len is not None else max_length
         self.tensor_parallel_size = int(tensor_parallel_size)
         self.data_parallel_size = int(data_parallel_size)
         self.model_args = {
@@ -74,6 +80,7 @@ please install vllm via `pip install lm-eval[vllm]` or `pip install -e .[vllm]`"
             "tokenizer_revision": tokenizer_revision,
             "trust_remote_code": trust_remote_code,
             "tensor_parallel_size": int(tensor_parallel_size),
+            "max_model_len": int(self._max_length) if self._max_length else None,
             "swap_space": int(swap_space),
             "quantization": quantization,
             "seed": int(seed),
@@ -89,7 +96,6 @@ please install vllm via `pip install lm-eval[vllm]` or `pip install -e .[vllm]`"
             tokenizer_revision=tokenizer_revision,
         )
         self.batch_size = batch_size
-        self._max_length = max_length
         self._max_gen_toks = max_gen_toks
 
     @property