add vllm model

lm_eval/models/vllm.py

@@ -74,7 +74,7 @@ class VLLM(LM):
 
     def _model_generate(
         self,
-        requests: List = None,
+        requests: List[int] = None,
         generate: bool = False,
         max_tokens: int = None,
         stop: Optional[List[str]] = None,
@@ -153,7 +153,7 @@ class VLLM(LM):
         # batch tokenize contexts
         context, all_gen_kwargs = zip(*(req.args for req in requests))
         context_encoding = self.tokenizer(context)
-        requests = zip((context, context_encoding.input_ids), all_gen_kwargs)
+        requests = list(zip((context, context_encoding.input_ids), all_gen_kwargs))
 
         def _collate_gen(_requests):
             # the negative sign on len(toks) sorts descending - this has a few advantages:
@@ -269,7 +269,7 @@ class VLLM(LM):
                 inps.append(inp)
                 ctxlens.append(ctxlen)
 
-            outputs = self._model_generate(generate=False)
+            outputs = self._model_generate(requests=inps, generate=False)
 
             for output, ctxlen, (cache_key, context_enc, continuation_enc) in zip(
                 outputs, ctxlens, chunk