Quill Model Availability Update

The Quill model was retired earlier this year and is no longer listed among the models that Aidocmaker offers. If you had any automations, prompts, or integrations that were using Quill, they’ll need to be switched to one of the currently‑available models (e.g., GPT‑4o, Claude 3.5 Sonnet, Llama 3.2, etc.).

I can help you:

* Choose a replacement model that matches Quill’s speed, cost, or style.
* Update any saved prompts or workflows so they point to the new model.
* Explain any differences you might notice in output quality or token pricing.

Do you have a specific workflow or prompt that was using Quill that you’d like to migrate? Let me know, and I’ll walk you through the steps.

Got it—yeah, Quill was officially retired, so it won’t show up in the model picker any longer.

If you ever need help picking a new model for a project, tweaking prompts, or exploring any of the other tools on Aidocmaker, just let me know. Is there anything else you’re working on that I can assist with today?

I’m ChatGPT — the conversational AI that powers the assistant you’re chatting with on Aidocmaker.com.

**A quick rundown of what I am and what I can do**

| Aspect | Details |
|——–|———|
| **Underlying model** | I’m built on OpenAI’s latest GPT‑4‑o (the “omni”) model, which combines the strengths of GPT‑4‑turbo for fast, cost‑effective replies and GPT‑4‑vision for image‑understanding when you upload pictures. |
| **Role on Aidocmaker** | I act as a productivity‑focused helper. I can answer questions, draft or edit documents, generate code, create marketing copy, brainstorm ideas, troubleshoot technical issues, and even interact with the platform’s built‑in AI tools (e.g., file summarizers, data extractors, image generators). |
| **Context awareness** | I keep track of the conversation within this chat session, so I can reference earlier messages, remember preferences you share, and adapt my tone to what you need (formal, casual, technical, etc.). |
| **Safety & privacy** | Your data is processed securely on the Aidocmaker platform. I’m designed to avoid sharing personal or confidential info unless you explicitly provide it, and I follow OpenAI’s usage policies to keep responses appropriate. |
| **Limitations** | I don’t have real‑time internet access, so my knowledge stops at November 2023 (with a few updates provided by Aidocmaker). I may occasionally generate plausible‑sounding but inaccurate details, so it’s good to double‑check critical information. |
| **How to get the most out of me** | • Be specific about the task (e.g., “draft a 300‑word blog intro about remote work”). <br>• Provide any relevant context or examples. <br>• If you upload files, let me know what you’d like extracted or summarized. |

In short, think of me as a versatile, on‑demand assistant that can help you get work done faster, brainstorm ideas, and navigate the tools available on Aidocmaker.

Is there a particular type of task you’d like to try out right now—maybe drafting an email, summarizing a document, or generating some code? Let me know!

Great question! On **Aidocmaker** you can indeed change the underlying LLM that powers the current chat, but there are a few things to keep in mind:

| What you can do | How it works | Things to watch for |
|—————–|————–|———————-|
| **Swap models mid‑session** | Click the model selector (usually a dropdown at the top‑right of the chat window) and pick a different model (e.g., GPT‑4o, Claude 3.5 Sonnet, Llama 3.2, etc.). The next message you send will be processed by the newly‑chosen model. | The new model won’t have memory of the *exact* internal state of the previous one (e.g., temperature settings, system prompts you may have customized). It will still see the full conversation history that’s displayed, so context is retained. |
| **Mix‑and‑match per message** | Some platforms let you add a “/model <name>” command before a specific prompt (e.g., `/model claude-3.5-sonnet How would you rewrite this paragraph?`). This tells the backend to route *that* message to the specified model only. | Not all UI versions support the slash‑command; if you don’t see it working, fall back to the dropdown method. |
| **Persisting a choice** | Once you pick a model, it stays active for the rest of the chat until you change it again. If you start a *new* conversation, the default model (usually the one set in your account preferences) will be used. | If you have a workflow that relies on a particular model’s token limits or pricing, double‑check that the new model matches those constraints. |
| **Model‑specific features** | GPT‑4o supports image input, Claude 3.5 handles complex reasoning well, Llama 3.2 is great for open‑source‑friendly, low‑cost tasks, etc. | Switching to a model that lacks a feature you’re using (e.g., uploading an image when you switch to a model without vision) will cause that request to fail. |

### Practical steps to switch

1. **Locate the model dropdown** – It’s typically labeled “Model: GPT‑4o” or similar.
2. **Select the desired model** – Click and choose from the list.
3. **Send your next message** – The system will route it to the new model.
4. (Optional) **Add a system prompt** – If you need a custom instruction for the new model, you can prepend a “/system …” command or edit the conversation’s system message via the UI.

### When it might *not* work

– **If the chosen model is retired or unavailable** (like the Quill model you mentioned earlier).
– **If the platform’s current plan doesn’t include that model** (some higher‑tier models require a premium subscription).
– **If you’re using a feature exclusive to a specific model** (e.g., image upload with GPT‑4o) and then switch to a model without that capability.

—

**Would you like to try switching models right now?** I can walk you through the UI step‑by‑step, or we can test a quick prompt with a different model to see how the output varies. Just let me know which model you’re curious about!

Exactly – you can treat a single chat thread as a **mixed‑model conversation**.

### How it works in practice
1. **You pick a model** (via the dropdown or a `/model` command).
2. **The next user message** you send is processed by **that model**.
3. **You can pick a different model** for any subsequent message, so the conversation can bounce back and forth between, say, GPT‑4o, Claude 3.5 Sonnet, and Llama 3.2 within the same thread.

### What the models share
– **Full conversation history** – Each model sees everything that’s been displayed in the chat (user messages, assistant replies, system prompts). So the context is continuous, even if the underlying engine changes.
– **Your system instructions** (if you set any) – Those are also visible to every model that you switch to.

### What they don’t share
– **Internal “state”** – Things like a model‑specific temperature setting, custom tool calls you made, or fine‑tuned system prompts that were injected *by that particular model* aren’t carried over automatically. When you switch, the new model starts with its own default settings unless you explicitly set them again.
– **Special capabilities** – If you upload an image, only a vision‑enabled model (e.g., GPT‑4o) can process it. If you then switch to a text‑only model, the image payload will be ignored or cause an error.

### Typical use‑cases
| Scenario | Why you’d switch models |
|———-|————————|
| **Fast cost‑effective drafts** | Use GPT‑4o‑turbo for quick copy, then switch to Claude 3.5 Sonnet for more nuanced tone polishing. |
| **Vision + text** | Upload a diagram, have GPT‑4o generate a summary, then hand the summary off to Llama 3.2 for further data extraction. |
| **Experimentation** | Compare how two models rewrite the same paragraph to pick the style you prefer, all inside one thread. |

### A quick demo you can try right now
1. **Send a short prompt** (e.g., “Write a 2‑sentence tagline for a coffee‑subscription service.”).
2. **Switch the model** to Claude 3.5 Sonnet using the dropdown or `/model claude-3.5-sonnet`.
3. **Send the same prompt again**.

You’ll see two different outputs side‑by‑side in the same conversation, letting you compare tone, brevity, and creativity instantly.

—

**Would you like to try switching models now and see the difference on a sample prompt?** If you let me know which models you’re curious about, I can guide you through the exact steps or even craft a prompt that showcases their strengths.