update docs

API.md

@@ -5,7 +5,7 @@ This allows you to easily integrate it with a different frontend of your choice.
 
 For full API documentation, see [localhost:8100/docs](localhost:8100/docs) after deployment.
 
-If you want to see the API docs before deployment, check out the [hosted docs here](https://opengpts-example-vz4y4ooboq-uc.a.run.app/docs).
+If you want to see the API docs before deployment, check out the [hosted docs here](https://opengpts-backend-ffoprvkqsa-uc.a.run.app/docs).
 
 In the examples below, cookies are used as a mock auth method. For production, we recommend using JWT auth. Refer to the [auth guide for production](auth.md) for more information.
 When using JWT auth, you will need to include the JWT in the `Authorization` header as a Bearer token.
@@ -40,14 +40,14 @@ The config parameters allows you to set the LLM used, the instructions of the as
   "config": {
     "configurable": {
       "type": "agent",
-      "type==agent/agent_type": "GPT 3.5 Turbo",
+      "type==agent/agent_type": "GPT 4o Mini",
       "type==agent/system_message": "You are a helpful assistant",
       "type==agent/tools": ["Wikipedia"]
   },
   "public": True
 }
 ```
-This creates an assistant with the name `"bar"`, with GPT 3.5 Turbo, with a prompt `"You are a helpful assistant"` using the Wikipedia tool , that is public.
+This creates an assistant with the name `"bar"`, with GPT 4o Mini, with a prompt `"You are a helpful assistant"` using the Wikipedia tool , that is public.
 
 Available tools names can be found in the AvailableTools class in backend/packages/gizmo-agent/gizmo_agent/tools.py
 Available llms can be found in GizmoAgentType in backend/packages/gizmo-agent/gizmo_agent/agent_types/__init__.py

README.md

@@ -29,7 +29,7 @@ Because this is open source, if you do not like those architectures or want to m
 
 **Key Links**
 
-- [GPTs: a simple hosted version](https://opengpts-example-vz4y4ooboq-uc.a.run.app/)
+- [GPTs: a simple hosted version](https://opengpts-backend-ffoprvkqsa-uc.a.run.app/)
 - [Assistants API: a getting started guide](API.md)
 - [Auth: a guide for production](auth.md)
 
@@ -281,7 +281,7 @@ the most flexible choice, but they work well with fewer models and can be less r
 
 When creating an assistant, you specify a few things.
 
-First, you choose the language model to use. Only a few language models can be used reliably well: GPT-3.5, GPT-4, 
+First, you choose the language model to use. Only a few language models can be used reliably well: GPT-4,
 Claude, and Gemini.
 
 Second, you choose the tools to use. These can be predefined tools OR a retriever constructed from uploaded files. You 
@@ -339,21 +339,20 @@ You can choose between different LLMs to use.
 This takes advantage of LangChain's many integrations.
 It is important to note that depending on which LLM you use, you may need to change how you are prompting it.
 
-We have exposed four agent types by default:
+We have exposed 3 agent types by default:
 
-- "GPT 3.5 Turbo"
-- "GPT 4"
-- "Azure OpenAI"
-- "Claude 2"
+- "GPT 4o Mini"
+- "GPT 4o"
+- "Claude 3.5 Sonnet"
 
 We will work to add more when we have confidence they can work well.
 
 If you want to add your own LLM or agent configuration, or want to edit the existing ones, you can find them in 
-`backend/app/agent_types`
+`backend/app/llms` (`LLMType`)
 
-#### Claude 2
+#### Claude 3.5 Sonnet
 
-If using Claude 2, you will need to set the following environment variable:
+If using Claude, you will need to set the following environment variable:
 
 ```shell
 export ANTHROPIC_API_KEY=sk-...