Enable AutoModelForCausalLM direct loading with Quanto FP8 quantization_config

config.json

@@ -29,5 +29,10 @@
   "tie_word_embeddings": false,
   "transformers_version": "4.57.6",
   "use_cache": true,
-  "vocab_size": 130560
-}
+  "vocab_size": 130560,
+  "quantization_config": {
+    "quant_method": "quanto",
+    "weights": "float8",
+    "activations": null
+  }
+}