OpenCode can now officially be used with your Grok or X Premium subscriptions Try it with the new Grok Build model

@opencode Anyone already tried out the rate limits? I can them pretty fast in grok.com So probably enough for a minute of coding? :)

@opencode @grok what’s context window size, knowledge cut-off date of Grok Build model?

@opencode This is actually useful. Stop making model access feel like a theology exam. If people are already paying for Grok, just let it into the damn toolchain.

@opencode Does not show up with any of the providers on my end.

@opencode Awesome, whats the usage limits with X Premium?

The Wienie 500 is back. Tune in at 2pm ET on FOX to see who will taste victory, cook the competition, and earn the title of top dog.

@opencode @grok if i use my X premium sub for this will I get charged per the tokens i burn in opencode?

@opencode Am I doing something wrong?

grok build for codebase intelligence is the part i'd want to stress-test first. our stream aggregator runs across 4 platform apis and most coding agents stall out once the context spans more than two integration patterns. speed only matters if it stays coherent way deeper than a handful of turns. curious how opencode session memory holds up when the grok call set spans the whole repo not just a single file

@opencode So how does this work, like I can only access the grok models ? Is it part of the go subscription? So do I BYOK?

@opencode wohoo, testing asap. just heard the news here on the ai news radio, fully run by ai btw :) x.com/thehypedotnews…

@opencode successfully connected and using it.

@opencode I have used opencdoe with x.Ai for some times. It needs to fix frequent oldString problem first. Such a brutal issue. Switch to qwen code cli, works much better(although once I faced with oldString issue, but it can fix itself and reread the file content).

@opencode Every AI company slowly becoming one giant VS Code extension

@opencode Anyone found where you can see usage?

@opencode it only asks me for an API key for this and Codex...

@opencode attaching X subscription to OpenCode – wasn't on my bingo card tbh Absolutely giving it a shot

@opencode Every AI company right now: ‘Why build your own IDE when we can plug directly into the one developers already use?’

@opencode grok build is harness, why it’s listed on the model selection i dunt quite understand

@opencode I tried Grok Build, I was surprise d by its coding ability, a bit better than Gemini 3.5 Flash

@opencode I've been dying to try out Grok Build. I still don't see the OAuth option, but at least I can use Zen now

Grok Imagine now has dramatically improved lip sync and sharper audio quality on all image-to-video generations. Dialogue tracks the mouth. Sound matches the scene. Your videos look and sound the way you imagined them. Try SuperGrok today.

@opencode @nexxeln It’s funny how a couple of days ago I was begging elon to let us use grok on opencode via the subscription model like the competitors are doing. Kinda happy it happened

@opencode does open code currently support app-server mode like codex or like claude -p mode?

@opencode waiting for qwen model got grok model , what are the limits btw ??

@opencode Is this X Premium or X Premium+ subs?

@opencode @grok what is this? How to use it productively?

@opencode premium? not premium+? no way $8 gonna give you more than 5 minutes of usage right?

@opencode Doesn't work with my non-heavy grok sub.

@opencode in opencode, after /connect, I only see xAI (which refers to the API) not supergrok. this is windows opencode version 1.15.6

Vibe coding your next app? Your AI agent can add production auth in minutes. Clerk Skills gives it the context to get auth right.

@opencode Why not just make Grok Build open source? This is like using those off-brand Lego-compatible "bricks" instead of Legos.

@opencode Who uses Grok for coding? 🤢🤮 … to the bin...

Grok subscribers just woke up with a coding agent they did not know they had yesterday. That is the smartest kind of distribution. No new audience to build. No new pricing to justify. Just plug into people who are already paying and already trust the ecosystem. OpenCode did not launch a product today. They found a shortcut to millions of users.

@opencode Smart move — bundling inference into existing subs lowers the activation energy massively. This is how you turn casual users into daily coders. Curious to test it on a real codebase migration.

@opencode 256K context + no output cap + included in an X subscription is kinda wild Feels like the AI subscription war is shifting from: “best model” to “best overall ecosystem value.”

@opencode It's good, they are not limiting it to x platform only.

@opencode LFG !!!