feat: use new model-proxy in production (#1908)

README.md

@@ -180,7 +180,7 @@ docker exec -it $(docker ps -q -f ancestor=letta/letta) letta run
 In the CLI tool, you'll be able to create new agents, or load existing agents:
 ```
 🧬 Creating new agent...
-? Select LLM model: letta-free [type=openai] [ip=https://inference.memgpt.ai]
+? Select LLM model: letta-free [type=openai] [ip=https://inference.letta.com]
 ? Select embedding model: letta-free [type=hugging-face] [ip=https://embeddings.memgpt.ai]
 ->  🤖 Using persona profile: 'sam_pov'
 ->  🧑 Using human profile: 'basic'
@@ -246,7 +246,7 @@ letta run
 ```
 ```
 🧬 Creating new agent...
-? Select LLM model: letta-free [type=openai] [ip=https://inference.memgpt.ai]
+? Select LLM model: letta-free [type=openai] [ip=https://inference.letta.com]
 ? Select embedding model: letta-free [type=hugging-face] [ip=https://embeddings.memgpt.ai]
 ->  🤖 Using persona profile: 'sam_pov'
 ->  🧑 Using human profile: 'basic'

letta/constants.py

@@ -4,7 +4,7 @@ from logging import CRITICAL, DEBUG, ERROR, INFO, NOTSET, WARN, WARNING
 LETTA_DIR = os.path.join(os.path.expanduser("~"), ".letta")
 LETTA_TOOL_EXECUTION_DIR = os.path.join(LETTA_DIR, "tool_execution_dir")
 
-LETTA_MODEL_ENDPOINT = "https://inference.memgpt.ai"
+LETTA_MODEL_ENDPOINT = "https://inference.letta.com"
 
 ADMIN_PREFIX = "/v1/admin"
 API_PREFIX = "/v1"

letta/llm_api/openai.py

@@ -158,7 +158,7 @@ def build_openai_chat_completions_request(
         # data.response_format = {"type": "json_object"}
 
     if llm_config.model_endpoint == LETTA_MODEL_ENDPOINT:
-        # override user id for inference.memgpt.ai
+        # override user id for inference.letta.com
         import uuid
 
         data.user = str(uuid.UUID(int=0))

letta/llm_api/openai_client.py

@@ -136,7 +136,7 @@ class OpenAIClient(LLMClientBase):
         )
 
         if llm_config.model_endpoint == LETTA_MODEL_ENDPOINT:
-            # override user id for inference.memgpt.ai
+            # override user id for inference.letta.com
             import uuid
 
             data.user = str(uuid.UUID(int=0))

tests/configs/letta_hosted.json

@@ -1,11 +1,11 @@
 {
-    "context_window": 8192,
-    "model_endpoint_type": "openai",
-    "model_endpoint": "https://inference.memgpt.ai",
-    "model": "memgpt-openai",
-    "embedding_endpoint_type": "hugging-face",
-    "embedding_endpoint": "https://embeddings.memgpt.ai",
-    "embedding_model": "BAAI/bge-large-en-v1.5",
-    "embedding_dim": 1024,
-    "embedding_chunk_size": 300
+  "context_window": 8192,
+  "model_endpoint_type": "openai",
+  "model_endpoint": "https://inference.letta.com",
+  "model": "memgpt-openai",
+  "embedding_endpoint_type": "hugging-face",
+  "embedding_endpoint": "https://embeddings.memgpt.ai",
+  "embedding_model": "BAAI/bge-large-en-v1.5",
+  "embedding_dim": 1024,
+  "embedding_chunk_size": 300
 }

tests/configs/llm_model_configs/letta-hosted.json

@@ -1,7 +1,7 @@
 {
   "context_window": 8192,
   "model_endpoint_type": "openai",
-  "model_endpoint": "https://inference.memgpt.ai",
+  "model_endpoint": "https://inference.letta.com",
   "model": "memgpt-openai",
   "put_inner_thoughts_in_kwargs": true
 }