Replies: 6 comments 2 replies
-
works fine here. |
Beta Was this translation helpful? Give feedback.
-
Tengo la misma duda/problema. |
Beta Was this translation helpful? Give feedback.
-
github copilot has gone worse, also it just doesnt remmember the currrent converstaion, its like everytime when i am interacting i need to start from scratch. is anyone facing the same issue? |
Beta Was this translation helpful? Give feedback.
-
🕒 Discussion Activity Reminder 🕒 This Discussion has been labeled as dormant by an automated system for having no activity in the last 60 days. Please consider one the following actions: 1️⃣ Close as Out of Date: If the topic is no longer relevant, close the Discussion as 2️⃣ Provide More Information: Share additional details or context — or let the community know if you've found a solution on your own. 3️⃣ Mark a Reply as Answer: If your question has been answered by a reply, mark the most helpful reply as the solution. Note: This dormant notification will only apply to Discussions with the Thank you for helping bring this Discussion to a resolution! 💬 |
Beta Was this translation helpful? Give feedback.
-
@Sascha-L This is important: The AI you are talking to does not know what it is! It's a machine learning algorithm trained to respond like an assistant would and nothing else. All it "knows" is its training data and the system prompt (this is a prompt inserted before your conversation to give the ai some context). It says that gpt-4 is unreleased (and gpt 4 would say the same), because at the time of training, there couldn't have been any information in the training data about it yet. You need to use AI with caution and not trust everything it says. |
Beta Was this translation helpful? Give feedback.
-
According to release notes, so far only the Chat extension can use gpt-4o. The inline suggestions are most likely still provided by codex. var DW = "copilot-codex",
hit = new ta(4, "customModels"),
SW = class SW {
async getModels(t) {
let n = await t.get(Va).getCopilotToken(t);
return new G6(n);
}
logModelsForToken(t, n) {
let i = new G6(n).getModelIds();
i.length > 1 && i.forEach((o) => hit.info(t, `Available model: ${o}`));
}
};
r(SW, "AvailableModelManager");
var Tf = SW,
FW = class FW {
constructor(t) {
this.token = t;
}
getModelIds() {
var n, a;
return [...((a = (n = this.token.getTokenValue("cml")) == null ? void 0 : n.split(",")) != null ? a : []), DW];
}
async getModelForResource(t, n, a = void 0) {
let i = zi(t, Tn.DebugOverrideEngine);
if (i) return new j2(i, DW === i);
if (a) {
let s = t.get(Vn).customEngine(a);
if (s) return new j2(s, DW === s);
}
let o = this.getModelIds();
return new j2(o[0]);
}
}; |
Beta Was this translation helpful? Give feedback.
-
Select Topic Area
Question
Body
Hi,
today I installed the latest version of the GitHub Copilot Chat extension to Visual Studio 2022 17.9.6. I singed up for a 30-day trial.
I played around with it and found out, that the results are always worse than the results from free version of ChatGPT 4.
When I asked GitHub Copilot Chat, which version it is using, the answer is: It is using GPT-3 from 2022 and GPT-4 is not released yet.
What's wrong?
Beta Was this translation helpful? Give feedback.
All reactions