chore: a few documentation enhancements for main README and FAQs

FAQ.md

@@ -2,6 +2,18 @@
 
 # bolt.diy
 
+## Recommended Models for bolt.diy
+
+For the best experience with bolt.diy, we recommend using the following models:
+
+- **Claude 3.5 Sonnet (old)**: Best overall coder, providing excellent results across all use cases
+- **Gemini 2.0 Flash**: Exceptional speed while maintaining good performance
+- **GPT-4o**: Strong alternative to Claude 3.5 Sonnet with comparable capabilities
+- **DeepSeekCoder V2 236b**: Best open source model (available through OpenRouter, DeepSeek API, or self-hosted)
+- **Qwen 2.5 Coder 32b**: Best model for self-hosting with reasonable hardware requirements
+
+**Note**: Models with less than 7b parameters typically lack the capability to properly interact with bolt!
+
 ## FAQ
 
 ### How do I get the best results with bolt.diy?
@@ -34,14 +46,18 @@ We have seen this error a couple times and for some reason just restarting the D
 
 We promise you that we are constantly testing new PRs coming into bolt.diy and the preview is core functionality, so the application is not broken! When you get a blank preview or don’t get a preview, this is generally because the LLM hallucinated bad code or incorrect commands. We are working on making this more transparent so it is obvious. Sometimes the error will appear in developer console too so check that as well.
 
-### How to add a LLM:
+### Everything works but the results are bad
 
-To make new LLMs available to use in this version of bolt.new, head on over to `app/utils/constants.ts` and find the constant MODEL_LIST. Each element in this array is an object that has the model ID for the name (get this from the provider's API documentation), a label for the frontend model dropdown, and the provider. 
+This goes to the point above about how local LLMs are getting very powerful but you still are going to see better (sometimes much better) results with the largest LLMs like GPT-4o, Claude 3.5 Sonnet, and DeepSeek Coder V2 236b. If you are using smaller LLMs like Qwen-2.5-Coder, consider it more experimental and educational at this point. It can build smaller applications really well, which is super impressive for a local LLM, but for larger scale applications you want to use the larger LLMs still!
 
-By default, Anthropic, OpenAI, Groq, and Ollama are implemented as providers, but the YouTube video for this repo covers how to extend this to work with more providers if you wish!
+### Received structured exception #0xc0000005: access violation
 
-When you add a new model to the MODEL_LIST array, it will immediately be available to use when you run the app locally or reload it. For Ollama models, make sure you have the model installed already before trying to use it here!
+If you are getting this, you are probably on Windows. The fix is generally to update the [Visual C++ Redistributable](https://learn.microsoft.com/en-us/cpp/windows/latest-supported-vc-redist?view=msvc-170)
 
-### Everything works but the results are bad
+### How to add an LLM:
 
-This goes to the point above about how local LLMs are getting very powerful but you still are going to see better (sometimes much better) results with the largest LLMs like GPT-4o, Claude 3.5 Sonnet, and DeepSeek Coder V2 236b. If you are using smaller LLMs like Qwen-2.5-Coder, consider it more experimental and educational at this point. It can build smaller applications really well, which is super impressive for a local LLM, but for larger scale applications you want to use the larger LLMs still!
+To make new LLMs available to use in this version of bolt.new, head on over to `app/utils/constants.ts` and find the constant MODEL_LIST. Each element in this array is an object that has the model ID for the name (get this from the provider's API documentation), a label for the frontend model dropdown, and the provider. 
+
+By default, many providers are already implemented, but the YouTube video for this repo covers how to extend this to work with more providers if you wish!
+
+When you add a new model to the MODEL_LIST array, it will immediately be available to use when you run the app locally or reload it.

README.md

@@ -4,7 +4,9 @@
 
 Welcome to bolt.diy, the official open source version of Bolt.new (previously known as oTToDev and bolt.new ANY LLM), which allows you to choose the LLM that you use for each prompt! Currently, you can use OpenAI, Anthropic, Ollama, OpenRouter, Gemini, LMStudio, Mistral, xAI, HuggingFace, DeepSeek, or Groq models - and it is easily extended to use any other model supported by the Vercel AI SDK! See the instructions below for running this locally and extending it to include more models.
 
-Check the [bolt.diy Docs](https://stackblitz-labs.github.io/bolt.diy/) for more information. This documentation is still being updated after the transfer.
+Check the [bolt.diy Docs](https://stackblitz-labs.github.io/bolt.diy/) for more information. 
+
+We have also launched an experimental agent called the "bolt.diy Expert" that can answer common questions about bolt.diy. Find it here on the [oTTomator Live Agent Studio](https://studio.ottomator.ai/).
 
 bolt.diy was originally started by [Cole Medin](https://www.youtube.com/@ColeMedin) but has quickly grown into a massive community effort to build the BEST open source AI coding assistant!
 
@@ -211,4 +213,4 @@ Explore upcoming features and priorities on our [Roadmap](https://roadmap.sh/r/o
 
 ## FAQ
 
-For answers to common questions, visit our [FAQ Page](FAQ.md).
+For answers to common questions, issues, and to see a list of recommended models, visit our [FAQ Page](FAQ.md).

docs/docs/FAQ.md

@@ -1,5 +1,19 @@
 # Frequently Asked Questions (FAQ)
 
+## What are the best models for bolt.diy?
+
+For the best experience with bolt.diy, we recommend using the following models:
+
+- **Claude 3.5 Sonnet (old)**: Best overall coder, providing excellent results across all use cases
+- **Gemini 2.0 Flash**: Exceptional speed while maintaining good performance
+- **GPT-4o**: Strong alternative to Claude 3.5 Sonnet with comparable capabilities
+- **DeepSeekCoder V2 236b**: Best open source model (available through OpenRouter, DeepSeek API, or self-hosted)
+- **Qwen 2.5 Coder 32b**: Best model for self-hosting with reasonable hardware requirements
+
+**Note**: Models with less than 7b parameters typically lack the capability to properly interact with bolt!
+
+---
+
 ## How do I get the best results with bolt.diy?
 
 - **Be specific about your stack**:  
@@ -72,6 +86,12 @@ Local LLMs like Qwen-2.5-Coder are powerful for small applications but still exp
 
 ---
 
+### **"Received structured exception #0xc0000005: access violation"**
+
+If you are getting this, you are probably on Windows. The fix is generally to update the [Visual C++ Redistributable](https://learn.microsoft.com/en-us/cpp/windows/latest-supported-vc-redist?view=msvc-170)
+
+---
+
 ### **"Miniflare or Wrangler errors in Windows"**
 You will need to make sure you have the latest version of Visual Studio C++ installed (14.40.33816), more information here https://github.com/stackblitz-labs/bolt.diy/issues/19.