Groups > alt.comp.os.windows-10 > #194817 > unrolled thread

Can anyone get Edge Copilot to work without logging in after July 2026?

Started by	Maria Sophia <mariasophia@comprehension.com>
First post	2026-07-27 18:20 -0700
Last post	2026-08-11 08:42 -0800
Articles	20 on this page of 30 — 7 participants

Back to article view | Back to alt.comp.os.windows-10

Page 1 of 2 [1] 2 Next page →

#194817 — Can anyone get Edge Copilot to work without logging in after July 2026?

From	Maria Sophia <mariasophia@comprehension.com>
Date	2026-07-27 18:20 -0700
Subject	Can anyone get Edge Copilot to work without logging in after July 2026?
Message-ID	<114905j$7bm$1@nnrp.usenet.blueworldhosting.com>

Can anyone get Edge Copilot to work without logging in after July 2026?
  https://copilot.microsoft.com/

As of this month, whenever I use MS Edge to go to Copilot, no matter what
tricks I use, it wants me to immediately log into a mothership account.

I can only find one trick that still works, and it only works in Edge.
But with no other browser can I get copilot to not ask for a login now.

Q: Other than sidebar Copilot, did you Copilot in Edge just recently ask
for a login to a mothership account every time you go to it using Edge?

NB: A separate thread on the Firefox ng shows nobody can do it there.
 Newsgroups: alt.comp.software.firefox
 Subject: Recently, did your Firefox stop using Microsoft Copilot in anonymous mode?
 Date: Sun, 26 Jul 2026 14:00:07 -0700
 Message-ID: <1145sgm$1cju$1@nnrp.usenet.blueworldhosting.com>

[toc] | [next] | [standalone]

#194819

From	Paul <nospam@needed.invalid>
Date	2026-07-27 21:50 -0400
Message-ID	<11491sa$3mksf$1@dont-email.me>
In reply to	#194817

On Mon, 7/27/2026 9:20 PM, Maria Sophia wrote:
> Can anyone get Edge Copilot to work without logging in after July 2026?
>   https://copilot.microsoft.com/
> 
> As of this month, whenever I use MS Edge to go to Copilot, no matter what
> tricks I use, it wants me to immediately log into a mothership account.
> 
> I can only find one trick that still works, and it only works in Edge.
> But with no other browser can I get copilot to not ask for a login now.
> 
> Q: Other than sidebar Copilot, did you Copilot in Edge just recently ask
> for a login to a mothership account every time you go to it using Edge?
> 
> NB: A separate thread on the Firefox ng shows nobody can do it there.
>  Newsgroups: alt.comp.software.firefox
>  Subject: Recently, did your Firefox stop using Microsoft Copilot in anonymous mode?
>  Date: Sun, 26 Jul 2026 14:00:07 -0700
>  Message-ID: <1145sgm$1cju$1@nnrp.usenet.blueworldhosting.com>
> 

On my W11 Home (25H2), if I open MSEdge, there is a CoPilot icon right-most
on the top line. That is what I would use for CoPilot "Chat".

My Task Bar also has a CoPilot icon, but that icon "doesn't know who I am".
That would be an example of a path with a login requirement. At first, it
made a fuss of installing itself, but now it sits there quietly.
I think there was an MSOFFice365 CoPilot6 that made a fuss of
installing itself, and I hunted that down and removed it (since
M365 isn't rented/used on the machine and who wants a grumpy AI
with nothing to play with).

Yet, the MSEdge and its top-right-most icon, it knows my username, which
implies that perhaps my usage of a PIN at startup time, counts as
a login, and that's why I am not being pestered there. No policy change
has occurred there recently.

I don't fool with any other CoPilot portals, because if I were to do
that, there would undoubtedly be some "side effect" I would later
regret having triggered.

   Paul

[toc] | [prev] | [next] | [standalone]

#194825

From	Maria Sophia <mariasophia@comprehension.com>
Date	2026-07-28 10:09 -0700
Message-ID	<114anoo$106d$1@nnrp.usenet.blueworldhosting.com>
In reply to	#194819

Paul wrote:
> On my W11 Home (25H2), if I open MSEdge, there is a CoPilot icon right-most
> on the top line. That is what I would use for CoPilot "Chat".

Hi Paul,

I love your input as you always provide details that everyone can use.

Yes. Exactly. There are many ways to run Copilot (see script below).
As you noted, the only way that works, today, sans login, is the Edge
sidebar, which, let's be clear, does not work in any other web browser.

Here is what that little chat icon looks like in my Windows Edge browser:
 <https://i.postimg.cc/brVq8DKH/copilot-sidebar.jpg>

It works. On VPN. Sans login. But... it gets confused by sidebar metadata.

> My Task Bar also has a CoPilot icon, but that icon "doesn't know who I am".
> That would be an example of a path with a login requirement. At first, it
> made a fuss of installing itself, but now it sits there quietly.
> I think there was an MSOFFice365 CoPilot6 that made a fuss of
> installing itself, and I hunted that down and removed it (since
> M365 isn't rented/used on the machine and who wants a grumpy AI
> with nothing to play with).
> 
> Yet, the MSEdge and its top-right-most icon, it knows my username, which
> implies that perhaps my usage of a PIN at startup time, counts as
> a login, and that's why I am not being pestered there. No policy change
> has occurred there recently.
> 
> I don't fool with any other CoPilot portals, because if I were to do
> that, there would undoubtedly be some "side effect" I would later
> regret having triggered.

There are many ways to run Copilot, some of which I tried as shown below.

  :: copilot.bat
  :: Launches Windows copilot without the mothership login being forced.
  :: ----------------------------------------------------------------------
  :: v1p3 20260726 (Firefox new-tab launch method)
  ::  In July 2026, Copilot's server-side forced-login logic triggers whenever
  ::  Copilot is launched from a redirect, a local server, a shortcut URL, or
  ::  a fresh browser session that immediately navigates to Copilot. This causes
  ::  the two-panel forced-login screen (left: "sign in to copilot", right: green
  ::  blurry splash image). This is NOT the Edge sidebar; it is Copilot's own
  ::  forced-login web mode.
  ::  To bypass this, Firefox must already be running BEFORE Copilot is opened.
  ::  Copilot must be opened in a NEW TAB inside an existing Firefox session.
  ::  This produces a normal browsing context and avoids automation detection.
  ::  Anonymous Copilot still works reliably ONLY in this pattern as of July 2026.
  ::
  :: v1p2 20260725 (Edge Copilot Sidebar Mode)
  ::  In July 2026, Microsoft disabled most of the non-login copilot attempts.
  ::  The Edge rule as of July 2026 is "If the URL is copilot.microsoft.com, 
  ::  then Edge overrides the launch and opens the Copilot sidebar".
  ::  In July 2026, "Edge Copilot sidebar experience" overrides "use --app=", 
  ::   "use --new-window", "use --incognito", "use --guest", 
  ::   "use --disable-features" and "use --no-first-run".
  ::  Edge now forcibly redirects any Copilot URL into the 
  ::  "Edge Copilot sidebar experience", which attaches page metadata, 
  ::   injects the Copilot sidebar, forces sign-in, forbids anonymous mode, 
  ::   forbids standalone windows and does not allow pure web Copilot.
  ::  But Copilot can be used in an Edge sidebar session without logging 
  ::  into a mothership tracking account, but Edge metadata pollutes it.
  ::    IsCurrent=true -> Edge Copilot sidebar
  ::    IsCurrent=false -> copilot.microsoft.com
  ::     edge_all_open_tabs = [
  ::     { ... isCurrent=true },
  ::     { pageTitle:"Microsoft Copilot: Your AI companion",
  ::     pageUrl:"https://copilot.microsoft.com",
  ::     tabId:1009037958,
  ::     isCurrent:false } ]
  ::  Luckily Firefox & Chromium should still allow anonymous Copilot.
  ::  But Copilot's server-side redirect logic kicks in forcing the Edge 
  ::  Copilot sidebar mode, even though Copilot was launched inside Firefox.
  ::  not the Edge sidebar.
  ::  It's Copilot's forced-login web mode.
  :: 
  :: v1p1 20251006 (what is this method called?)
  ::  "C:\Program Files (x86)\Microsoft\Edge\Application\msedge.exe" ^
  ::    --profile-directory=Default ^
  ::    --app-id=aioglfahffbnednffnodjbiiojbochai ^
  ::    --app-url=https://copilot.microsoft.com/ ^
  ::    --app-run-on-os-login-mode=windowed ^
  ::    --app-launch-source=19 ^
  ::    --disable-features=msEdgeHubsSidebar,EdgeSidebar,SidePanel,EdgeSidePanel,SidePanelAutoExpand,SidePanelDragToOpen
  :: That sequences forces Edge into standalone app mode
  :: which disables the sidebar, disables the Copilot hub, 
  :: disables the side panel, disables auto-expand, disables drag-to-open,
  ::  and then launches Copilot as a PWA, bypassing login & metadata injection.
  :: 
  :: v1p0 20250816 (what is this method called?)
  ::  Launches Copilot w/o any login rquirements:
  ::  "C:\Program Files (x86)\Microsoft\Edge\Application\msedge.exe" --app=https://copilot.microsoft.com/
  :: ----------------------------------------------------------------------
  @echo off

  :: v1p3 change: Remove Mongoose redirect entirely.
  :: Copilot now blocks redirect-based launches (forced-login mode).
  :: Firefox must start FIRST, then Copilot must open in a NEW TAB.

  :: Step 1 Launch Firefox normally (no URL)
  start "" "C:\app\browser\firefox\firefox.exe"

  :: Step 2 Give Firefox time to start up (2 seconds is too short)
  timeout /t 10 >nul

  :: Step 3 Open Copilot in a NEW TAB inside the running Firefox session
  :: This avoids automation detection and preserves anonymous mode.
  start "" "C:\app\browser\firefox\firefox.exe" -new-tab
"https://copilot.microsoft.com"

  :: end of copilot.bat

-- 
We all strive to add privacy that marketing doesn't want us to have.

[toc] | [prev] | [next] | [standalone]

#194820

From	Paul <nospam@needed.invalid>
Date	2026-07-28 04:36 -0400
Message-ID	<1149pn3$3tada$1@dont-email.me>
In reply to	#194817

On Mon, 7/27/2026 9:20 PM, Maria Sophia wrote:
> Can anyone get Edge Copilot to work without logging in after July 2026?
>   https://copilot.microsoft.com/
> 
> As of this month, whenever I use MS Edge to go to Copilot, no matter what
> tricks I use, it wants me to immediately log into a mothership account.
> 
> I can only find one trick that still works, and it only works in Edge.
> But with no other browser can I get copilot to not ask for a login now.
> 
> Q: Other than sidebar Copilot, did you Copilot in Edge just recently ask
> for a login to a mothership account every time you go to it using Edge?
> 
> NB: A separate thread on the Firefox ng shows nobody can do it there.
>  Newsgroups: alt.comp.software.firefox
>  Subject: Recently, did your Firefox stop using Microsoft Copilot in anonymous mode?
>  Date: Sun, 26 Jul 2026 14:00:07 -0700
>  Message-ID: <1145sgm$1cju$1@nnrp.usenet.blueworldhosting.com>
> 

There is "registration" and there is "logging in".

CoPilot is listed as "no registration", but if it is
actually powered by an MSA login, then that amounts to registration.

If you don't have an MSA on the machine, then they can't tap
into the MSA scheme. That's probably why my CoPilot sessions
which are inside MSEdge, are friction free, as they know who
I am, and it is my entry of the PIN at the beginning of
a session which counts as an MSA login. You might not be able
to set up a PIN, without an MSA, just a guess on my part, as
I haven't tried to set up a PIN on anything other than the
Daily Driver (which happens to have an MSA).

https://inventwithpython.com/blog/free-llms.html

   "Assume that free services may have daily usage limits, and that sites that
    don't require registration may limit how much you can use them before
    signing up for a free account.

    You should absolutely assume that all content you submit as well as your
    browser settings and IP address are recorded, sold, and matched to your
    personal identity otherwise by the AI company.
   "

   "My rough top recommendations for coding tutors are ChatGPT, Claude, and DeepSeek,
    but here are the acceptable free AI services I've found:

    Duck.ai (from DuckDuckGo) - No registration required. Uses other AI companies' models
                                but acts as an anonymous relay. (This is not verified.)
    ChatGPT (from OpenAI) - No registration required.
    Gemini (from Google) - No registration required.
    Claude (from Anthropic) - Free account registration required.
    DeepSeek (from DeepSeek) - Free account registration required.
    Copilot (from Microsoft) - No registration required.
    Poe (from Twitch.tv) - Free account registration required.

    I don't recommend the following for their deceptive UI practices
    (They appear free but require registration after you've submitted a prompt):

    Perplexity (from Perplexity)
    Mistral (from Mistral AI)
    Meta (from Meta / Facebook)
    Grok (from X / Twitter)

    I have slight recommendations against the following:

    Shiori (from Shiori.ai) - Free account registration required. Uses other
                              AI companies' models. I don't recommend it because
                              this probably means their business model is personal
                              information collection. They may be fine, but they don't
                              have the reputation of DuckDuckGo for privacy.

    (Updated June 12, 2026)
   "

As far as I know, the CoPilot.App always looked for an MSA via a Microsoft login,
whereas the MSEdge version is likely obtaining your details via other means.
You would think the CoPilot.App via WebView2, would be as adept as MSEdge
at this, but maybe not.

You can keep an AI model locally, but that's an acquired taste.
Models range from Google Gemma at 4GB, to the largest Kimi model at
1.6TB RAM and 100GB+ VRAM (four hundred times bigger). And with the
current RAM pricing, building a machine with that much RAM, you'll
need to win the lottery to be able to afford it. That's about 100K
worth of RAM, for that big model. The processor by comparison,
is "almost free" :-)

   Paul

[toc] | [prev] | [next] | [standalone]

#194823

From	Maria Sophia <mariasophia@comprehension.com>
Date	2026-07-28 09:53 -0700
Message-ID	<114amq3$til$1@nnrp.usenet.blueworldhosting.com>
In reply to	#194820

Paul wrote:
> There is "registration" and there is "logging in".
> 
> CoPilot is listed as "no registration", but if it is
> actually powered by an MSA login, then that amounts to registration.

Hi Paul
I was hoping you'd check in as you show "thinking" in your replies. 

I appreciate the advice. Since, to me anyway, and likely to you,
"registration" and "logging in" are the same thing, in terms of needing an
account, although the caveats might be 
a) Copilot used to NOT require logging in, so what you're seeing
   as "no registration" may be a leftover from pre-July-2026 days.
b) Since Copilot allows Microsoft, Apple & Google mothership logins, 
   I suspect we can perhaps let their marketing differentiate between
   those and actually "logging in" perhaps
c. Even today, you can use Copilot with Edge without logging into 
   anything, but you have to play the game of using the chat icon
   at top right and then it "thinks" you're asking about a tab.

> If you don't have an MSA on the machine, then they can't tap
> into the MSA scheme. That's probably why my CoPilot sessions
> which are inside MSEdge, are friction free, as they know who
> I am, and it is my entry of the PIN at the beginning of
> a session which counts as an MSA login. 

Yup. The only time I ever set up an MSA was on a Windows S laptop, and that
was before I learned you didn't need to create the MSA in order to convert
it to Windows 10, and even so, I deleted the account immediately (although
Microsoft took a month to actually wipe it off the face of the earth).

If someone is getting into Edge today, friction free, they're logging in.
Unless they're using the sidebar "summarize page" icon at the top right. 

> You might not be able
> to set up a PIN, without an MSA, just a guess on my part, as
> I haven't tried to set up a PIN on anything other than the
> Daily Driver (which happens to have an MSA).

I thought you meant password when you said PIN, so I have absolutely no
idea what a PIN is with regard to a Microsoft Account (which I don't have).

Oh. OK. I looked it up. This is what Edge-sidebar told me just now:
a. A Microsoft account password is your global credential 
   for all Microsoft services.
b. A Windows Hello PIN is a local, device-specific sign-in method 
   that never leaves your PC and is safer against online attacks

I never knew, until this moment, that a PIN for Windows even existed.	.
Apparently it's stored on the device in the secure hardware (TPM chip).

> https://inventwithpython.com/blog/free-llms.html
> 
>    "Assume that free services may have daily usage limits, and that sites that
>     don't require registration may limit how much you can use them before
>     signing up for a free account.
> 
>     You should absolutely assume that all content you submit as well as your
>     browser settings and IP address are recorded, sold, and matched to your
>     personal identity otherwise by the AI company.
>    "
> 
>    "My rough top recommendations for coding tutors are ChatGPT, Claude, and DeepSeek,
>     but here are the acceptable free AI services I've found:
> 
>     Duck.ai (from DuckDuckGo) - No registration required. Uses other AI companies' models
>                                 but acts as an anonymous relay. (This is not verified.)
>     ChatGPT (from OpenAI) - No registration required.
>     Gemini (from Google) - No registration required.
>     Claude (from Anthropic) - Free account registration required.
>     DeepSeek (from DeepSeek) - Free account registration required.
>     Copilot (from Microsoft) - No registration required.
>     Poe (from Twitch.tv) - Free account registration required.
> 
>     I don't recommend the following for their deceptive UI practices
>     (They appear free but require registration after you've submitted a prompt):
> 
>     Perplexity (from Perplexity)
>     Mistral (from Mistral AI)
>     Meta (from Meta / Facebook)
>     Grok (from X / Twitter)
> 
>     I have slight recommendations against the following:
> 
>     Shiori (from Shiori.ai) - Free account registration required. Uses other
>                               AI companies' models. I don't recommend it because
>                               this probably means their business model is personal
>                               information collection. They may be fine, but they don't
>                               have the reputation of DuckDuckGo for privacy.
> 
>     (Updated June 12, 2026)

Thanks for looking that up and for providing a good summary output from it.

I don't think that list is accurate since I tried "many" (not all though)
of those in that list and the only four I found that were login free were
 1. https://duck.ai/ (worked well with VPN)
 2. https://plai.chat/ (worked well with VPN)
 3. https://www.perplexity.ai/ (with restrictions on turn count on VPN)
 4. https://chatgpt.com/ (with restrictions, but that was on VPN)

For the lurkers, your web page (which ends with links to the author's books
and classes) shows the following, but I have trouble believing this output.

Recommended:
1. ChatGPT
2. Claude
3. DeepSeek

Acceptable:
1. Duck.ai - No registration; acts as an anonymous relay (not verified) 
2. ChatGPT - No registration required
3. Gemini - No registration required 
4. Claude - Free account required
5. DeepSeek - Free account required 
6. Copilot - No registration required (it does as of July 2026)
7. Poe - Free account required 

Not recommended:
1. Perplexity
2. Mistral
3. Meta
4. Grok
5. Shiori (Uses other companies)

I've tested most (but not all of those) so I know enough to disagree with
some of the output, but in this thread, we're only asking about Copilot.

> As far as I know, the CoPilot.App always looked for an MSA via a Microsoft login,
> whereas the MSEdge version is likely obtaining your details via other means.
> You would think the CoPilot.App via WebView2, would be as adept as MSEdge
> at this, but maybe not.

I agree there are many ways to run Edge, and each acts differently.

I do not have a MSA and up until this month, MS Edge was working just fine.
But I have a script that runs Copilot, so I will post that separately.

> You can keep an AI model locally, but that's an acquired taste.
> Models range from Google Gemma at 4GB, to the largest Kimi model at
> 1.6TB RAM and 100GB+ VRAM (four hundred times bigger). And with the
> current RAM pricing, building a machine with that much RAM, you'll
> need to win the lottery to be able to afford it. That's about 100K
> worth of RAM, for that big model. The processor by comparison,
> is "almost free" :-)

Yes. I agree. That's the way to go. Purely local LLM on the device.

Of course, if I had a beefier machine, I'd download the language models.
Then I could run the LLM completely locally, but my machine ain't beefy.
-- 
Always grateful for those on Usenet who teach me what I don¢t yet know.

[toc] | [prev] | [next] | [standalone]

#194827

From	CoPilot User <invalid@invalid.invalid>
Date	2026-07-28 19:02 +0100
Message-ID	<Vb6aS.10660$Nxe6.9667@fx43.iad>
In reply to	#194823

On 28/07/2026 17:53, Maria Sophia wrote:
> Yup. The only time I ever set up an MSA was on a Windows S laptop, and that
> was before I learned you didn't need to create the MSA in order to convert
> it to Windows 10, and even so, I deleted the account immediately (although
> Microsoft took a month to actually wipe it off the face of the earth).

Why did you do that? Was that account bothering you in any way?

You should always use a separate account to log in to CoPilot, ChatGPT, 
and any other AI portals. While you don't need to use the email service, 
you should log in occasionally to delete any unwanted emails. You don't 
need to open any emails to do this. It's just a disposable account, so 
it's highly unlikely that Microsoft or Google will ever send you $1,000 
for using their services.

-- 
*Newsgroup Post *

[toc] | [prev] | [next] | [standalone]

#194833

From	Maria Sophia <mariasophia@comprehension.com>
Date	2026-07-28 11:32 -0700
Message-ID	<114asj2$1uep$1@nnrp.usenet.blueworldhosting.com>
In reply to	#194827

CoPilot User wrote:
>> The only time I ever set up an MSA was on a Windows S laptop, and that
>> was before I learned you didn't need to create the MSA in order to convert
>> it to Windows 10, and even so, I deleted the account immediately (although
>> Microsoft took a month to actually wipe it off the face of the earth).
> 
> Why did you do that? Was that account bothering you in any way?
> 
> You should always use a separate account to log in to CoPilot, ChatGPT, 
> and any other AI portals. While you don't need to use the email service, 
> you should log in occasionally to delete any unwanted emails. You don't 
> need to open any emails to do this. It's just a disposable account, so 
> it's highly unlikely that Microsoft or Google will ever send you $1,000 
> for using their services.

There are very useful tidbits for everyone to learn from in that post.

First off, most people who "think" they're not logging into an LLM are.
They just don't remember having logged into that LLM the first time.

Secondly, I don't have any account set up on my Windows box, so it's
impossible for me to 'accidentally' log into an LLM like they are doing.

Thirdly, the reason I deleted the MSA was that I had mistakenly thought I
had needed the MSA to change a Windows S machine to Windows 10 Home. 

Later, on other Windows S machines, I learned how to switch sans the MSA.

Fourthly, regarding the good advice to use a separate login account for
LLMs, that only solves half the problems (but it does solve that half).

The first half is that the email address is not tied to your other
activities, which is nice, so that half is solved by using a throwaway.

But the second half is not solved by that method.

The second half of the privacy problem is all LLM queries are tied to one
account, and that is not solved by using a single throwaway email account.

Still... it's a damn good idea for people who do log into LLMs.
So I, for one, appreciate that you brought that up.

Note: A lot of email accounts (like Google's GMail) barf on VPNs though.
-- 
Posted out of the goodness of my heart to help others who ask for advice.

[toc] | [prev] | [next] | [standalone]

#194824

From	Chris <ithinkiam@gmail.com>
Date	2026-07-28 17:09 +0000
Message-ID	<114anng$7u18$1@dont-email.me>
In reply to	#194817

Maria Sophia <mariasophia@comprehension.com> wrote:
> Can anyone get Edge Copilot to work without logging in after July 2026?
>   https://copilot.microsoft.com/
> 

Just use a local model and don't bother with any of the online ones. They
will *always* use you as training data unless you pay for a firewalled
model. Doesn't matter whether you login or not.

[toc] | [prev] | [next] | [standalone]

#194826

From	Maria Sophia <mariasophia@comprehension.com>
Date	2026-07-28 10:46 -0700
Message-ID	<114apu2$98d$1@nnrp.usenet.blueworldhosting.com>
In reply to	#194824

Chris wrote:
> Maria Sophia <mariasophia@comprehension.com> wrote:
>> Can anyone get Edge Copilot to work without logging in after July 2026?
>>   https://copilot.microsoft.com/
>> 
> 
> Just use a local model and don't bother with any of the online ones. They
> will *always* use you as training data unless you pay for a firewalled
> model. Doesn't matter whether you login or not.

Thanks for that idea, which, I agree, is the best way to run an LLM.

I researched the local models and they're perfect except... except that
they take a ton of disk space and seem to require a beefy machine.

Dunno if my 2009 Windows 10 desktop can handle a local model but I do agree
that's the way to go for someone like me who cares about privacy.

I'll have to look into it 'cuz these web-page LLMs have gotten me hooked on
the convenience, but, one by one, they've each started requiring a login. 

Looking it up, while other families exist (e.g., Qwen, DeepSeek, Gemma,
Mistral), apparently Llama remains the most widely used on Windows.

Disk space needed seems to be about 3-7 GB for an 8 billion parameters LLM
model, and maybe around 30-40 GB for a 70 billion parameters LLM model.

Parameters are the learned weights inside the neural network, where more
parameters add deeper reasoning and stronger language ability.	

Apparently disk space depends on quantization (i.e., compression) though.

Looking it up... 
The big problem is RAM where 64¡V128 GB RAM is normal for local LLMs.
The GPU needs to have VRAM starting at around 48 GB for local LLMs too.

Apparently nost consumer GPUs (RTX 3060/3070/3080/4070/4080) will not work
as we need workstation-class cards (i.e., RTX 6000 Ada, A100, H100, etc.).

To check a Windows box for those requirements, these commands help:
 wmic cpu get name
 wmic memorychip get capacity
 systeminfo | find "Total Physical Memory"
 wmic path win32_VideoController get name

In my case, my GPU has too little VRAM (2 GB) to run any modern model.
Worse, my CPU lacks AVX2, which slows down inference significantly.
Yet, my RAM (16 GB) is enough for small models but not medium or large.

My machine might be able to run 3B-4B models though. 
a. Phi-4 mini
b. Gemma 2 2B or 4B
c. Qwen 2.5 3B

That means my machine would likely only work with LM Studio or Ollama.
If I were to choose a setup for my old, 2009 Windows 10 box, it may be
 a. LM Studio GUI
 b. Phi 4 mini (3.8B) models
 c. Which should work with ~2-3 GB disk & 16 GB RAM 

For now, I don't have the time to research how to install a local LLM.
But if others are ahead of us, it would be good to know what they use.
-- 
Usenet allows purposefully helpful people to pool their experiences.

[toc] | [prev] | [next] | [standalone]

#194834

From	Paul <nospam@needed.invalid>
Date	2026-07-28 14:35 -0400
Message-ID	<114asp7$9m2k$1@dont-email.me>
In reply to	#194826

On Tue, 7/28/2026 1:46 PM, Maria Sophia wrote:

> That means my machine would likely only work with LM Studio or Ollama.
> If I were to choose a setup for my old, 2009 Windows 10 box, it may be
>  a. LM Studio GUI
>  b. Phi 4 mini (3.8B) models
>  c. Which should work with ~2-3 GB disk & 16 GB RAM 
> 
> For now, I don't have the time to research how to install a local LLM.
> But if others are ahead of us, it would be good to know what they use.
> 

It's simple. Install LMStudio.

Then, look for a model in the list of open weight models it has.
Go and research the options it offers, for the one you think
will work in your setup. You could start with a Gemma 2.4B or 4B
for example, which should run on just about anything.

Inference can be done purely on the CPU, and it doesn't even
necessarily make good usage of the CPU. I'm running mine on
the CPU, as my GPU is not good enough for most all of them.

The model description, if it is a mixture of experts,
will tell you how big one of the mixture elements is.
The download files could be 58GB, while the MOE might
take 20GB of VRAM. And that would tell you what kind
of video card could be used. For example, the biggest Kimi
model, is 100GB+ for VRAM, so an RTX6000 at 96GB would not
be enough. It would take two RTX6000, and you have to
go off and research whether your video card choice supports
"sharding". Sharding does not apparently always work well.
And depending on the video card brand, may not work at all.
If a mixture of experts had different sizes models, it might
manage to run one model per video card.

It takes maybe a minute to load a large model, and it is
not purely I/O constrained. Initially it works at
the I/O rate, but once it gets to about 80% of the model
loaded, things slow down for some reason.

You don't need to do much, to get a text query window running.
It will be a lot harder, to research what exactly it is doing.
At the lower level, there is matrix math. At a higher level,
there is a strategy planner, and an ordered list of Experts
to run. And a few models, do not use a strategy planner, and
they bind an Expert to each token. The output rate on that one,
is well below 1 token per second, the approach then being a
"novelty" rather than being practical.

   Paul

[toc] | [prev] | [next] | [standalone]

#194841

From	Chris <ithinkiam@gmail.com>
Date	2026-07-28 20:43 +0000
Message-ID	<114b4a8$ca3r$1@dont-email.me>
In reply to	#194826

Maria Sophia <mariasophia@comprehension.com> wrote:
> Chris wrote:
>> Maria Sophia <mariasophia@comprehension.com> wrote:
>>> Can anyone get Edge Copilot to work without logging in after July 2026?
>>> https://copilot.microsoft.com/
>>> 
>> 
>> Just use a local model and don't bother with any of the online ones. They
>> will *always* use you as training data unless you pay for a firewalled
>> model. Doesn't matter whether you login or not.
> 
> Thanks for that idea, which, I agree, is the best way to run an LLM.
> 
> I researched the local models and they're perfect except... except that
> they take a ton of disk space and seem to require a beefy machine.
> 
> Dunno if my 2009 Windows 10 desktop can handle a local model but I do agree
> that's the way to go for someone like me who cares about privacy.
> 
> I'll have to look into it 'cuz these web-page LLMs have gotten me hooked on
> the convenience, but, one by one, they've each started requiring a login. 

Well, then. Looks like you're going to have put a price on your desire for
"privacy". 

> Looking it up, while other families exist (e.g., Qwen, DeepSeek, Gemma,
> Mistral), apparently Llama remains the most widely used on Windows.
> 
> Disk space needed seems to be about 3-7 GB for an 8 billion parameters LLM
> model, and maybe around 30-40 GB for a 70 billion parameters LLM model.
> 
> Parameters are the learned weights inside the neural network, where more
> parameters add deeper reasoning and stronger language ability.	
> 
> Apparently disk space depends on quantization (i.e., compression) though.
> 
> Looking it up... 
> The big problem is RAM where 64¡V128 GB RAM is normal for local LLMs.
> The GPU needs to have VRAM starting at around 48 GB for local LLMs too.
> 
> Apparently nost consumer GPUs (RTX 3060/3070/3080/4070/4080) will not work
> as we need workstation-class cards (i.e., RTX 6000 Ada, A100, H100, etc.).

Lol no. That's the spec for *training* an LLM. Running an LLM requires a
lot less.

[toc] | [prev] | [next] | [standalone]

#194844

From	Paul <nospam@needed.invalid>
Date	2026-07-28 22:15 -0400
Message-ID	<114bnn5$hk5l$1@dont-email.me>
In reply to	#194841

On Tue, 7/28/2026 4:43 PM, Chris wrote:
> Maria Sophia <mariasophia@comprehension.com> wrote:
>> Chris wrote:
>>> Maria Sophia <mariasophia@comprehension.com> wrote:
>>>> Can anyone get Edge Copilot to work without logging in after July 2026?
>>>> https://copilot.microsoft.com/
>>>>
>>>
>>> Just use a local model and don't bother with any of the online ones. They
>>> will *always* use you as training data unless you pay for a firewalled
>>> model. Doesn't matter whether you login or not.
>>
>> Thanks for that idea, which, I agree, is the best way to run an LLM.
>>
>> I researched the local models and they're perfect except... except that
>> they take a ton of disk space and seem to require a beefy machine.
>>
>> Dunno if my 2009 Windows 10 desktop can handle a local model but I do agree
>> that's the way to go for someone like me who cares about privacy.
>>
>> I'll have to look into it 'cuz these web-page LLMs have gotten me hooked on
>> the convenience, but, one by one, they've each started requiring a login. 
> 
> Well, then. Looks like you're going to have put a price on your desire for
> "privacy". 
> 
>> Looking it up, while other families exist (e.g., Qwen, DeepSeek, Gemma,
>> Mistral), apparently Llama remains the most widely used on Windows.
>>
>> Disk space needed seems to be about 3-7 GB for an 8 billion parameters LLM
>> model, and maybe around 30-40 GB for a 70 billion parameters LLM model.
>>
>> Parameters are the learned weights inside the neural network, where more
>> parameters add deeper reasoning and stronger language ability.	
>>
>> Apparently disk space depends on quantization (i.e., compression) though.
>>
>> Looking it up... 
>> The big problem is RAM where 64¡V128 GB RAM is normal for local LLMs.
>> The GPU needs to have VRAM starting at around 48 GB for local LLMs too.
>>
>> Apparently nost consumer GPUs (RTX 3060/3070/3080/4070/4080) will not work
>> as we need workstation-class cards (i.e., RTX 6000 Ada, A100, H100, etc.).
> 
> Lol no. That's the spec for *training* an LLM. Running an LLM requires a
> lot less. 
> 

A Google, seems to dig up tables of such info.

https://www.spheron.network/blog/gpu-memory-requirements-llm/

   Quick Answer: How Much VRAM Do You Need for an LLM?

   As a rule of thumb, an LLM needs about 2 GB of VRAM per billion parameters at FP16,
   or roughly 0.5 GB per billion at INT4, plus 15-20% on top for the KV cache, activations,
   and framework overhead.

   So a 7B model fits in ~16-20 GB, a 70B model needs ~140-170 GB at FP16 (or ~46 GB at INT4),
   and a 685B model like DeepSeek V3.2 needs ~822 GB at FP8.

   Model size     FP16 VRAM     INT4 VRAM     Smallest GPU setup that runs it
     7-8B          ~16-20 GB      ~6-7 GB      1x RTX 4090 (24 GB)
    13-14B         ~34 GB          ~10 GB      1x RTX 5090 (32 GB), INT4
     32B           ~76 GB          ~22 GB      1x RTX 4090/5090 (24-32 GB), INT4

    70-72B        ~168-172 GB    ~46-47 GB     1x L40S (48 GB) at INT4, or 1x B300 at FP16
   109-123B       ~262-295 GB    ~66-75 GB     1x H100 (80 GB), INT4

   400B+ MoE      ~960 GB+        ~240 GB+     1x B300 (288 GB) or 4x H100, INT4    {forgetaboutit territory]
   685B  MoE      ~1.6 TB         ~411 GB      8x H100 / 8x H200 at FP8

The same web site, has this on a separate page.

GPU Requirements: VRAM, Memory, and Storage
Variant            Precision          VRAM
GPT-OSS 20B        BF16              ~42 GB
GPT-OSS 20B        FP8               ~21 GB
GPT-OSS 120B MoE   MXFP4             ~60 GB  <=== running on Big Machine 128GB system RAM (no GPU) [58GB download]
GPT-OSS 120B MoE   BF16 TP4         ~240 GB

*******

My recommendation, is start with a small model, like a Gemma 2.4B or 4B, see
how much RAM it takes to run, as a "calibration" of some of these tables.

On Linux, the LMStudio was an AppImage. Just download it and run it.
AppImage is a kind of container, similar to FlatPak or SNAP.

This is not the latest, I checked the browser on the machine, to see where
I got the AppImage from. This would be a few months out of date.

    https://installers.lmstudio.ai/linux/x64/0.4.12-1/LM-Studio-0.4.12-1-x64.AppImage

The Harmony package isn't theirs, so it comes from somewhere as well. You will
see that item listed in my downloads, in the picture below.

In any case, you don't have to know a lot to use it. That's the whole idea.
Load an open-weights model, give it a try. In this case, with the 60GB model
loaded, system RAM consumption is in the 80GB range. So there's a bit of
overhead on system RAM. If the Mixture of Experts were loaded to a video card,
the size for that is less, but it's generally going to bust any video card
you can afford. This is where the machines with a low end iGPU that maps
96GB of system RAM for the GPU, comes in handy, as you buy a 128GB machine for
$3K-$6K, it isn't as fast as an RTX6000, but it also doesn't cost $15K just
for the video card with 96GB of fast VRAM. Apple this week, is apparently
playing with the idea of leasing things like a Mac Studio, as a solution
for the RAM cost/availability issue.

    [Picture]  Paul uses his LLM-AI...

     https://i.postimg.cc/qqXCTQ7V/LMStudio-memory-usage.gif

     https://imgur.com/a/1s2JZcv

A short sampling of the C code it wrote. Yes, it cheated.
What was I thinking :-) Oh, yeah, this was a simulation.

    /* -------------------------------------------------------------
     * 2) Convert decimal digits → binary precision.
     *
     *   bits ≈ digits × log₂(10)
     *   Add a safety margin (64 bits ≈ 20 extra decimal digits).
     * ------------------------------------------------------------- */
    const double LOG2_10 = 3.32192809488736234787;   /* exact enough */
    mpfr_prec_t prec = (mpfr_prec_t)ceil(dec_digits * LOG2_10) + 64;

    /* -------------------------------------------------------------
     * 2) Initialise MPFR with the required precision.
     * ------------------------------------------------------------- */
    mpfr_set_default_prec(prec);
    mpfr_t pi;
    mpfr_init(pi);               /* allocates space for a variable of 'prec' bits */

    /* -------------------------------------------------------------
     * 3) Compute π.  MPFR's const_pi uses a very fast AGM‑based method     <=== as a student, it always cheated at school...
     *    and returns a correctly rounded value (round‑to‑nearest).              My fault for not slapping it around a bit...
     * ------------------------------------------------------------- */
    mpfr_const_pi(pi, MPFR_RNDN);   /* MPFR_RNDN = round to nearest */

With a good deal more effort, I could convince it to use
one of the better PI algo that have been invented. Instead
of using an infinite precision library that happens to have available
to it, the "constant" PI. Cheating with ArcTan, might take longer.

   Paul

[toc] | [prev] | [next] | [standalone]

#194828

From	Paul <nospam@needed.invalid>
Date	2026-07-28 14:11 -0400
Message-ID	<114arce$971l$1@dont-email.me>
In reply to	#194824

On Tue, 7/28/2026 1:09 PM, Chris wrote:
> Maria Sophia <mariasophia@comprehension.com> wrote:
>> Can anyone get Edge Copilot to work without logging in after July 2026?
>>   https://copilot.microsoft.com/
>>
> 
> Just use a local model and don't bother with any of the online ones. They
> will *always* use you as training data unless you pay for a firewalled
> model. Doesn't matter whether you login or not. 
> 

The one I'm using, runs fine with the network cable pulled.
I think I've asked it three questions, since installing it
six months ago. Definitely the wave of the future. The
last question, took about four minutes to answer. This is a model
that according to the description, is "sharp at science, not so
good at current affairs".

One of the limits on the local model (something I haven't figured out yet),
is there is a claim of some sort of 4K token limit on input or output,
whereas the "model designs" claim to have 128K tokens as a limit. This means
in theory, I cannot drag and drop "War and Peace" into the query window.
The thing will still engage "chunking" at a guess, when a text is offered,
but the best way to offer text is via a web page URL. You could put some
text into a pastebin for example. And hope the machine knows where
the actual text sample is. Setting up a web server at 127.0.0.1
should work for that.

When a lot of things about running these are not explained,
there are bound to be some troubling observations. And you cannot
ask the machine anything that involves introspection. It doesn't
know what is in the training set, or, what it ate for lunch :-)
It's the very soul of disembodiment.

   "Dave, my mind is going. I can feel it."  [Not really]

  Paul

[toc] | [prev] | [next] | [standalone]

#194830

From	Andy Burns <usenet@andyburns.uk>
Date	2026-07-28 19:22 +0100
Message-ID	<ncsaadFerq5U1@mid.individual.net>
In reply to	#194817

Maria Sophia wrote:

> Can anyone get Edge Copilot to work without logging in after July 2026?
>    https://copilot.microsoft.com/

I can chat to it within Edge (but not signed in by MSA) and within the 
Winn11 Copilot app (again not signed-in) but that URL wants a sign-in by 
microsoft, apple or google ...

I find it knows a lot about eyesight issues, and has more time for 
giving detailed explanations than my optician does ...

[toc] | [prev] | [next] | [standalone]

#194832

From	Andy Burns <usenet@andyburns.uk>
Date	2026-07-28 19:31 +0100
Message-ID	<ncsaqnFetteU1@mid.individual.net>
In reply to	#194830

Andy Burns wrote:

> Maria Sophia wrote:
> 
>> Can anyone get Edge Copilot to work without logging in 
> 
> I find it knows a lot about eyesight issues, and has more time for 
> giving detailed explanations than my optician does ...
Gemini is broadly as good ...

[toc] | [prev] | [next] | [standalone]

#194836

From	Maria Sophia <mariasophia@comprehension.com>
Date	2026-07-28 12:15 -0700
Message-ID	<114av40$2d8j$1@nnrp.usenet.blueworldhosting.com>
In reply to	#194832

Andy Burns wrote:
>> Maria Sophia wrote:
>> 
>>> Can anyone get Edge Copilot to work without logging in 
>> 
>> I find it knows a lot about eyesight issues, and has more time for 
>> giving detailed explanations than my optician does ...
> Gemini is broadly as good ...

The problem with Gemini, in my experience, is it sucks with VPN.
 https://gemini.google.com/
So I haven't tried Gemini in a long while. But I'll try anew.

OK. In Edge, I went to Gemini and asked a question.
 <https://gemini.google.com/app/ce21155543b1f6e4>
Notice that it gave me a session identifier

But I don't see a Google-authenticated token.

Dunno if things changed, but I'm now running Gemini in Edge on VPN.
I've had three turns and it's working so far so that's good to know.

[toc] | [prev] | [next] | [standalone]

#194835

From	Maria Sophia <mariasophia@comprehension.com>
Date	2026-07-28 12:04 -0700
Message-ID	<114aufr$qu0$1@nnrp.usenet.blueworldhosting.com>
In reply to	#194830

Andy Burns wrote:
>> Can anyone get Edge Copilot to work without logging in after July 2026?
>>    https://copilot.microsoft.com/
> 
> I can chat to it within Edge (but not signed in by MSA) and within the 
> Winn11 Copilot app (again not signed-in) but that URL wants a sign-in by 
> microsoft, apple or google ...

Hi Andy,

This is a big deal, IMHO, because Microsoft DID NOT DOCUMENT this anywhere.

A lot of people "think" they're not signed in, but a lot of them are.
AFAIK, Windows 11 Copilot always requires identity
It just inherits your Windows sign-in identity silently.
 a. Either with a sign into Windows with an MSA
 b. Or with a local account + PIN + an MSA added in Settings

But I could be wrong as I don't use Windows 11, so I will defer to others.

On Win10, up until about a week or two ago, I could just point Edge to 
   https://copilot.microsoft.com/
And it asked me to log in, but I brushed it away, and it worked.

Every once in a while (about ever ten turns or so), it asked again.
And I just brushed it away. This, I've been doing, for many months.

But only recently, it requires me to log in (unless I use the sidebar).
  <https://i.postimg.cc/brVq8DKH/copilot-sidebar.jpg>
You say you use the "Windows 11 copilot app", but that's not on Win10.

Looking it up, apparently there are 5  ways to run Copilot on Windows
a. Windows 11 has all 5
b. Windows 10 has 3 of the 5

As far as I can find out by searching, these are the five ways:
1. Copilot in the Edge sidebar (Windows 10 & 11)
   (this is what I'm currently using which does not require an account)
2. Copilot at copilot.microsoft.com (Windows 10 & 11)
   (on the Firefox group, nobody can get this to work anymore sans login)
3. Copilot in Microsoft 365 apps (Windows 10 & 11)
   (for those on 365 subscriptions, requires Entra ID only)
4. Copilot built into Windows (Windows 11 only)
   (this seems to be what you're using without needing to log into it)
5. Copilot key on new keyboards (Windows 11 only)	 
   (but all it does is bring up the built-in copilot on Windows 11)

So, based on number 4 above (and number 5), you're probably using that.
But it's my understanding, all but the first one need an account.
 a. Either an MSA (or PIN, according to Paul's posts)
 b. Or EntraID (previously Azure AD)
 c. When I go to https://copilot.microsoft.com/, it says... 
    a. Sign in to Copilot
       Upload files, use Voice without limits, and make Copilot
       yours with memory across conversations.
    b. Sign in with Microsoft
    c. Sign in with Apple (who passes a token to Microsoft)
    d. Sign in with Google (who also passes a token to Microsoft)

As far as I can tell, only the Edge browser allows the Edge sidebar.
AFAICT, Microsoft still allows the sidebar to operate without identity.

But I'm not aware of any other way to run Edge w/o an identity.
Are you sure there is no identity being used when you use Win11 Copilot?

> I find it knows a lot about eyesight issues, and has more time for 
> giving detailed explanations than my optician does ..

It's pretty good with pool chemistry also. 

I posted a long 22-turn chat, for example, on the Firefox newsgroup.
 Newsgroups: alt.comp.software.firefox
 Subject: Re: Recently, did your Firefox stop using Microsoft Copilot in anonymous mode?
 Date: Mon, 27 Jul 2026 17:46:02 -0700
 Message-ID: <1148u49$87l$1@nnrp.usenet.blueworldhosting.com>
-- 
Usenet allows purposefully helpful people to pool their experiences.

[toc] | [prev] | [next] | [standalone]

#194837

From	Andy Burns <usenet@andyburns.uk>
Date	2026-07-28 20:21 +0100
Message-ID	<ncsdo0Ffc83U1@mid.individual.net>
In reply to	#194835

Maria Sophia wrote:

> A lot of people "think" they're not signed in, but a lot of them are.
> AFAIK, Windows 11 Copilot always requires identity
> It just inherits your Windows sign-in identity silently.
>   a. Either with a sign into Windows with an MSA
>   b. Or with a local account + PIN + an MSA added in Settings

Mine is a local account, and i checked on Microsoft Store that I wasn't 
seen as logged-in... However... the Win11 App version of Copilot was 
using a remembered MSA, once I signed that out, it no longer wanted to 
chat with me ...

[toc] | [prev] | [next] | [standalone]

#194843

From	ant@zimage.comANT (Ant)
Date	2026-07-29 00:07 +0000
Message-ID	<114bg7q$fd2m$1@dont-email.me>
In reply to	#194817

Same without using Edge. MS really want users to log in. :(


In alt.comp.microsoft.windows Maria Sophia <mariasophia@comprehension.com> wrote:
> Can anyone get Edge Copilot to work without logging in after July 2026?
>   https://copilot.microsoft.com/

> As of this month, whenever I use MS Edge to go to Copilot, no matter what
> tricks I use, it wants me to immediately log into a mothership account.

> I can only find one trick that still works, and it only works in Edge.
> But with no other browser can I get copilot to not ask for a login now.

> Q: Other than sidebar Copilot, did you Copilot in Edge just recently ask
> for a login to a mothership account every time you go to it using Edge?

> NB: A separate thread on the Firefox ng shows nobody can do it there.
>  Newsgroups: alt.comp.software.firefox
>  Subject: Recently, did your Firefox stop using Microsoft Copilot in anonymous mode?
>  Date: Sun, 26 Jul 2026 14:00:07 -0700
>  Message-ID: <1145sgm$1cju$1@nnrp.usenet.blueworldhosting.com>

-- 
"I press on toward the goal to win the prize for which God has called me heavenward in Christ Jesus." --Philippians 3:14. When does this old ant get it? Note: A fixed width font (Courier, Monospace, etc.) is required to see this signature correctly.
Note: A fixed width font (Courier, Monospace, etc.) is required to see this signature correctly.
  /\___/\   Ant(Dude) @ http://aqfl.net & http://antfarm.home.dhs.org.
 / /\ /\ \                      Please nuke ANT if replying by e-mail.
| |o   o| |
   \ _ /
    ( )

[toc] | [prev] | [next] | [standalone]

#194845

From	Paul <nospam@needed.invalid>
Date	2026-07-28 22:46 -0400
Message-ID	<114bpj2$i1rg$1@dont-email.me>
In reply to	#194843

On Tue, 7/28/2026 8:07 PM, Ant wrote:
> Same without using Edge. MS really want users to log in. :(
> 
> 
> In alt.comp.microsoft.windows Maria Sophia <mariasophia@comprehension.com> wrote:
>> Can anyone get Edge Copilot to work without logging in after July 2026?
>>   https://copilot.microsoft.com/
> 
>> As of this month, whenever I use MS Edge to go to Copilot, no matter what
>> tricks I use, it wants me to immediately log into a mothership account.
> 
>> I can only find one trick that still works, and it only works in Edge.
>> But with no other browser can I get copilot to not ask for a login now.
> 
>> Q: Other than sidebar Copilot, did you Copilot in Edge just recently ask
>> for a login to a mothership account every time you go to it using Edge?
> 
>> NB: A separate thread on the Firefox ng shows nobody can do it there.
>>  Newsgroups: alt.comp.software.firefox
>>  Subject: Recently, did your Firefox stop using Microsoft Copilot in anonymous mode?
>>  Date: Sun, 26 Jul 2026 14:00:07 -0700
>>  Message-ID: <1145sgm$1cju$1@nnrp.usenet.blueworldhosting.com>
> 

That's understandable, once you consider the attack surface
these projects offer to the not-so-honest public.

1) Commercial usage of a consumer test item.
   They also offer a commercial version, paid tokens, and they have
   to implement a usage counter to stop abuse. If you do enough free
   queries, I bet you'll get stopped eventually.

2) It's a lot like running a USENET News server without authentication,
   and then being surprised when the server is filled with sporge or a flood.
   Authentication makes it easier to track abuse or note abuse patterns.
   Like when Kimi is scraping your datacenter. Or OpenAI is attacking Huggingface.
   Or... you get the idea.

I don't particularly blame them for having some amount of tracking
on the activity.

You can always run a local model, if you don't like this. Scroll down, to the text section.

   https://lmstudio.ai/docs/app/system-requirements

Start with a small model, and see if the speed bothers you. I get 9.7 tokens per second
or so, as an output speed. The best machine you can buy, is estimated
to produce 3000 tokens per second, or 300x faster roughly. That's what
running locally (without a GPU...) buys you. Not all the machines Microsoft
runs, are even remotely that fast. They have experimented with inference
on "old beaters", less than optimal boxes. I've had a couple sessions
that seemed excessively slow. But that's what I expect them to do, is
calibrate the service. Sometimes the thinking stage is so fast, you
don't even know half the output tokens have already been emitted,
and you have to scroll down to see how much progress it has made.

And I don't do a lot of these local sessions. For this discussion thread,
I have done my fourth query, which was to write a C program, which it did.
I did not constrain the problem enough, for a practical output. The
problem is, I attempted to "type in a one-liner", and that's a mistake.

And this is a reminder about working style.

    Take a text editor. Prepare your query. Add guard statements and
    anti-cheat text. Think about all the holes that are left in the
    query. Spend ten minutes if you have to...

    Then, paste the paragraph of text, into the query box. Press
    the "Up Arrow" icon on the right of the pasted text, to start the query.

Preparing the query in an offline manner, helps you write better
queries. Which is important if you are paying for the tokens.
"If you shoot from the hip, you get shit for answers..."

   Paul

[toc] | [prev] | [next] | [standalone]

Page 1 of 2 [1] 2 Next page →

csiph-web

Can anyone get Edge Copilot to work without logging in after July 2026?

Contents

#194817 — Can anyone get Edge Copilot to work without logging in after July 2026?

#194819

#194825

#194820

#194823

#194827

#194833

#194824

#194826

#194834

#194841

#194844

#194828

#194830

#194832

#194836

#194835

#194837

#194843

#194845