Community
486 community available in the CoPilot directory
MiniMax M2.5 is now available in GitHub Copilot for free and student use
GitHub Copilot now lets Free users and students use MiniMax M2.5, but no official clients—Copilot CLI, VS Code, github.com/copilot UI, or OpenCode—display it. However, it's in the model listing API 3 times and and usable via GitHub OAuth. The model listing API provides a lot of detail about it. It indicates that M2.5 is being served by Fireworks, and that there's a "(Fast)" version served by Cerebras. Both are only available to "free" and "edu" subscriptions, to compensate, one would think, for the removal of premium models from the student plan ([https://github.com/orgs/community/discussions/189268).](https://github.com/orgs/community/discussions/189268)) Both are marked as preview, "model\_picker\_enabled=false" and "model\_picker\_category=powerful". The ID of "MiniMax M2.5 (Fast)" is `minimax-m2p5-cb` instead of `minimax-m2p5-fw`. { "billing": { "is_premium": true, "multiplier": 1, "restricted_to": ["free", "edu"] }, "capabilities": { "family": "minimax-m2p5-fw", "limits": {"max_context_window_tokens": 196608, "max_output_tokens": 32000, "max_prompt_tokens": 164000}, "object": "model_capabilities", "supports": { "parallel_tool_calls": true, "reasoning_effort": ["low", "medium", "high"], "streaming": true, "structured_outputs": true, "tool_calls": true }, "tokenizer": "o200k_base", "type": "chat" }, "id": "accounts/msft/routers/mp3yn0h7", "is_chat_default": false, "is_chat_fallback": false, "model_picker_category": "powerful", "model_picker_enabled": false, "name": "MiniMax M2.5 (Copilot)", "object": "model", "policy": {"state": "enabled", "terms": ""}, "preview": true, "supported_endpoints": ["/chat/completions"], "vendor": "Fireworks", "version": "accounts/msft/routers/mp3yn0h7" } The reason the official clients don't show the MiniMax models is because the `model_picker_enabled` field on them is set to false. In Piebald we weren't aware of that field, and therefore by accident we don't respect it, thus revealing these options. They must be pretty new because we have plenty of Copilot users and none of have used MiniMax with Copilot before.
D
Dramatic_Squash_3502 158
Is this a joke? weekly rate limit of 264 hrs (11 days)?
https://preview.redd.it/zflkaaam44vg1.png?width=462&format=png&auto=webp&s=658a1bae15dcf7428b6c993b2349fa7db4a858d1 Whats going on copilot team? **\[Update 1\]:** Tried after an hour - "Wait for 264 hrs" **\[Update 2\]:** Another try after some time - "Wait for 278 hrs" **\[Update 3\]:** Another try after some time - "Wait for 286 hrs" **\[Update 4\]:** After almost 18hrs - "Wait for 293 hrs" Copilot team wants me to take some break and go on a long vacation.. (hope I'd be sponsored) **\[Update 5 - 5:45pm IST\]:** "Wait for 301 hrs" And whats interesting is that for every such message response it costs credits, so i am trying with 0.33x models to say "hi" LOL **\[Update 6 - 11:00pm IST\]:** "You have reached your weekly rate limit. Please upgrade your plan or wait for your limit to reset on April 28" \-> wow weekly limit resets after 2 weeks. should rephrase the msg. \----------------- Usage is spread across multiple models, even including gemini 3.1, Sonnet but mainly gpt 5.4 and well within limits (15% so far should have been 50% so already way less usage), context managed as much possible, short tasks. So what they are saying has nothing to do with its, I believe its all about the shortage of their resources as the number of users have increased. And its a BUG, you cant have 300 hrs wait time for weekly limit .. or wait is it monthly limit.. but surely 15% usage wont be for monthly limit as well.
O
Ok-Cranberry4090 70
Pro+ Github blocked me for 2 days because I used their service
I own Github Pro+ Since February and it all was working fine. Not not only they removed Opus Fast, and "reduced limits on users" to only increase limit and now I cannot do work for 2 days? I specially reviewed their ToS that they so often quote for me to check, THERE IS NOTHING IN THERE that says rate limit, Github maybe you will enlighten me how can I know If I am going too fast so that I can Plan for this? I got 71% of requests left for the month, I won't be able to use them probably...
M
Muchaszewski 96
A long-awaited Feature is here!
GitHub Copilot CLI now supports remote sessions! https://github.blog/changelog/2026-04-13-remote-control-cli-sessions-on-web-and-mobile-in-public-preview/
M
mabdelhafiz94 109
I built a transparent Mario game that runs OVER my IDE so I can play while waiting for Copilot to write code.
When you're vibe coding and letting AI do the heavy lifting, the downtime gets boring fast, and continuous work is a real pain in the ass. To get a little relaxation, I built a stress-relief mini-game that lives directly on your Windows desktop as a transparent overlay, letting you stomp Goombas right on top of your open IDE. You can download the setup file from my GitHub (https://github.com/bxf1001g/desktop_mario/releases) and let it run silently in the background; just hit Ctrl+Alt+Shift+M to instantly bring the game up, press Esc to quickly hide it when you need to look busy, and hit Ctrl+Alt+Shift+M again whenever you need another break. This isn't the full version yet, just something fun I threw together as a side thing, so contributors are always welcome!
M
maxwellwatson1001 268
CoPilot Pro + VSCode extension is kinda a better deal than I expected: so far I vastly prefer GPT 5.4 Extra High to Claude Opus 4.6 and I'm only at 6% usage after at least like five hours of heavy work with it
CoPilot Pro + VSCode extension is kinda a better deal than I expected: so far I vastly prefer GPT 5.4 Extra High to Claude Opus 4.6 and I'm only at 6% usage after at least like five hours of heavy work with it
Z
ZootAllures9111 172
Goodbye trial accounts
https://github.blog/changelog/2026-04-10-pausing-new-github-copilot-pro-trials/ Free stuff is over. Personally I think this a great idea. If you cant afford ten bucks for some machine to do your thinking, you're doing it wrong. Rate limiting might go away if all the leeches are removed.
E
ElGuaco 80
AI is making mediocre engineers harder to spot
Not a hot take. Just something I’ve been noticing lately. Everyone on my team uses AI now. Code, infra, debugging, even architecture ideas. Productivity is definitely up. But… there’s a weird side effect. \--- Case 1 — trying everything, fixing nothing A guy was debugging a slow endpoint. Asked AI → got a bunch of suggestions: \- add caching \- batch requests \- async processing He tried all of them. Still slow. Turned out the query was missing an index. That’s it. The problem wasn’t that AI was wrong. It just wasn’t the right question. And if you don’t even know “missing index” is a thing to check, you’re basically guessing — just faster. \--- Case 2 — sounds right, breaks in real life Another one: someone built a rate limiter based on AI suggestions. AI said: “store counters in memory for performance”. Which… yeah, makes sense. Until you deploy multiple instances and everything falls apart. Now your rate limit is basically random. Again, AI didn’t lie. It just didn’t know (or wasn’t told) the real constraints. \--- That’s the pattern I keep seeing AI doesn’t make engineers worse. It just makes it easier to: \- look like you know what you’re doing \- ship something that “seems fine” \- and completely miss the actual problem \--- The scary part? These people look productive. \- PRs are clean \- features ship fast \- infra “works” But ask one level deeper: \- why this approach? \- what’s the trade-off? \- what happens under load? …and things get very quiet. \--- To be clear — I use AI every day I’m not anti-AI at all. It’s insanely good at: \- boilerplate \- exploring options \- explaining stuff quickly \- getting you unstuck But it’s not the one: \- making the final call \- understanding your system \- taking responsibility when things break That’s still on you. \--- Feels like the bar is shifting Before: \- you had to know stuff to build things Now: \- you can build things without fully understanding them And that gap only shows up when: \- something breaks \- or someone asks the “why” questions \--- If there’s one thing I’m trying to avoid right now: Becoming someone who can ship fast… but can’t think deeply. \--- Anyway, curious if others are seeing the same thing Is AI actually making us better engineers? Or just faster ones?
G
Ghost_Alpha- 58
Stricter rate limits coming... :(
Stricter rate limits coming... :(
J
JustARandomPersonnn 110
Anyone using the new Agents app?
Anyone using the new Agents app?
Z
zxyzyxz 72
share your best practice with github copilot for best results ?
I recently finished my project which crossed 50,000 lines of code (a review and discovery platform), and I managed to keep my total AI billing to just $120. I’ve been experimenting heavily with the current GitHub Copilot ecosystem and wanted to share the "stack" that worked for me—**while also hearing what you guys are using.** 1. The Model Selection: Planning vs. Execution I’ve stopped using one model for everything. My current workflow is: • Planning: I use Claude 4.6 Opus for the heavy architectural planning • Execution: I rely on codex Question for you: Are you sticking to one model, or are you finding better results splitting "thinking" and "typing" between models? 2. UI & Frontend: What’s ur procedure ? • What are you using for high-fidelity UI requests? 3.The MCP Game I’ve started integrating MCP (Model Context Protocol) servers to give my agent more "eyes" on the project. So far, i ised github mcp and mongodb and shadcn mcp server and better auth skills and also shadcn skills would share your experience guys
F
FitCoach5288 65
Message - Thanks to those who create Copilot
I want to thank everyone who makes Copilot!!! And also express my gratitude to Burke Holland for his awesome videos - clear, to the point, and no fluff. Join to me and say thank you to those who make our beloved Copilot!
C
ConsiderationIcy3143 120
Why is GitHub Copilot so affordable? Will it stay this way?
Hi everyone, I recently subscribed (after Cursor, TRAE and Antigravity) to GitHub Copilot Pro+ ($39/month) and I’m able to make a lot of requests without running out until the end of the month. It feels very generous compared to other AI services. Does anyone know why Copilot is priced so competitively? Do you think this pricing and usage model will stay the same long‑term, or should we expect changes in the future — for example, moving to token‑based usage? Just curious if this is here to stay or if major changes might be coming. Thanks!
W
WMPlanners 81
My experience using Copilot CLI alongside Claude Code at work — notes on the shared-skill problem
Sharing some notes from helping roll out Copilot CLI at my company, in case anyone else is in a similar mixed-tool situation. Context: different teams at my company landed on different AI coding tools. Some went with Copilot CLI (mostly the folks already deep in the GitHub ecosystem and using Copilot in VS Code), others went with Claude Code. As the person responsible for our shared internal tooling, I needed a way to write a skill once and have it work in both — otherwise we'd end up maintaining two copies of everything forever. A few things I ran into while working with Copilot CLI specifically: * **Plugin marketplace docs were harder to find than I expected.** Once I did find them, the format was reasonable, but the metadata file path and structure differ in subtle ways from what other tools use, so you can't just copy-paste a layout from elsewhere. * [**SKILL.md**](http://SKILL.md) `name` **field gotcha:** I had been prefixing names with a namespace (something like `myorg/skillname`) because that's conventional in other ecosystems. Copilot CLI silently fails to load the skill when you do that — no error, it just doesn't show up. Took me a while to track down. Plain names only. * **Copilot CLI + VS Code Agent Plugins:** Once the marketplace is set up correctly, installing from it inside VS Code worked smoothly, which was honestly the nicest part of the whole experience for the non-CLI people on the team. * **The** `version` **field behaves differently depending on where it appears** in the metadata hierarchy, which tripped me up more than once. My workaround for the cross-tool problem was to keep a single source of truth for skill metadata and use symlinks to expose it in the shape each tool expects. It took some trial and error to get the relative paths right, but once it clicked, editing one file updates both tools at once. I put the whole thing into a free template repo in case it helps anyone doing the same rollout — it's MIT licensed, no paid tier, and you can generate your own copy via "Use this template" on GitHub. Mostly though I'm curious whether others running Copilot CLI in a multi-tool environment have hit the same issues, and how you've handled it. The namespace-prefix thing in particular felt like something that should be documented more loudly. Repo if useful: [https://github.com/s950329/ai-plugin-skeleton](https://github.com/s950329/ai-plugin-skeleton) *(English isn't my first language, so I had AI help me clean up the writing — but the experience and the repo are mine.)*
S
s950329 45
Copilot CLI now supports BYOK and local models
Copilot CLI now let us connect our own model provider or run fully local models instead of using GitHub-hosted model routing. https://github.blog/changelog/2026-04-07-copilot-cli-now-supports-byok-and-local-models/
M
mabdelhafiz94 96
Rubber-duck agents are in Copilot now
Today, I have noticed some rubber-duck agents appearing in my tasks. It turns out it's a new feature of Copilot. Check it out at https://github.blog/ai-and-ml/github-copilot/github-copilot-cli-combines-model-families-for-a-second-opinion/
M
Much_Middle6320 65
I built a kanban board that auto-triggers Copilot agent teams by drag and drop
I love using copilot, but wanted to run agent teams using just one hand. To do this I built a VS Code plugin that triggers any terminal CLI agent, including Copilot, from just dragging around kanban cards. So if you have Copilot registered in the Lead Coder slot, when you drag a plan card there, it automatically sends a prompt to Copilot asking it to implement the plan. This also has really helped take advantage of Copilot's per-prompt pricing, since when you drag multiple cards into a column, it batches them all together and instructs Copilot to engage its subagents. When you have an agent in the planner slot, it also scores plans by complexity, so will auto direct work to the most appropriate terminal. Opus in Lead Coder slot, Sonnet in Coder Slot, and something cheap in the Intern slot. Used in this way, I'm able to knock over 5-10 low complexity tasks per Copilot credit, which really helps extend the quota. If you're interested you can find it (Switchboard) on any VS Code marketplace, I just released it today. It's fully open source as well, so you can vet it for security issues before installing: [https://github.com/TentacleOpera/switchboard](https://github.com/TentacleOpera/switchboard)
T
TheTentacleOpera 58
This many lines of code for $0.04
Feels like a good value? Especially considering how many different steps it took. I used Haiku 4.5 to generate the plan ($0.01) and 5.3-codex to execute ($0.03). $0.04 representing 0.4% of pro request allowance at $10.00/month. I finally bit the bullet and started the pro free-trial yesterday, as I'd exhausted every free allowance on every account on every provider I had, and it was only the 5th day of the month.
N
NotArticuno 189
300 Requests a month is a goldmine.
I have been a GitHub copilot member for more than 2 years and always used it for lightweight questions, fixes etc. now that all other providers are heavily limiting the uses, I am finding the 300 requests of GPT 5.4 Xhigh absolutely useful. I have been wasting this for so long.
P
PropperINC 107
Spec-driven Development with Copilot
SDD is great for vibe coding, but I always hit my quota with Claude Code or Codex before I even finish the planning phase. And Copilot is the answer.
M
Much_Middle6320 72
How I'm feeling as copilot power user
How I'm feeling as copilot power user
E
exitcactus 412
What are the advantages of using Copilot CLI over VS Code?
I'm thinking about giving the CLI a try. What are the advantages of Copilot CLI compared to the VS Code version?
H
hyperdx 105
gpt 5.4 mini is EXTREMELY request efficient
I use gpt 5.3 codex for the research/plan phase and use 5.4 mini to execute. it will use like .5% max even for huge refactors/changes in terms of planning it is kinda dumb even on high reasoning so use a different model for it. but with a detailed plan, it is REALLY good for execution. quite fast as well
S
snowieslilpikachu69 52
Claude Opus 4.6 extremely slow
In the past few days, I’ve noticed a massive slowdown with Claude Opus 4.6. The response speed has become painfully slow, sometimes reaching around **1 second per word**, which makes it almost unusable for longer outputs. I tested Opus 4.6 in "fast" mode, and interestingly, the speed now feels identical to how normal Opus 4.6 used to perform *before this degradation*. So it doesn’t really feel "fast" anymore, just baseline. My suspicion is that this might be due to a new **rate limiting mechanism** or some kind of throttling being applied recently. The drop in performance feels too consistent to be random lag. I'm in Pro+ plan.
T
tight_angel 52