Grok 4.3 has 1M context window with pricing of $1.25 input per million tokens

@techdevnotes It would be really helpful if we could see a meter showing how much voice time and context window we have left. Running into hard limits with no warning isn't great. A simple usage indicator would make a big difference. @grok @xai

@techdevnotes @grok how is the context window compared to other widely used AIs in the market?

@techdevnotes This is still the same 500B model we’ve been using via the web and app for a while now right?

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 it today in the Grok app.

@techdevnotes @grok whats the swe bench pro score

@techdevnotes And reduced pricing for the other models too. This is awesome!

@techdevnotes That’s awesome pricing.

@techdevnotes I expected 4.3 to be in expert and heavy now and 4.4 1t model to be coming out in beta. They need to move quicker. 4.3 is still a small model.

@techdevnotes Better than 4.20's pricing, but it's sad to see the pivot away from the low prices that 4.1 Fast had.

@techdevnotes Sonnet 4.6 level but with a strange personality. We're getting Grok 4.4 in a few weeks and I'm hoping to see that beat Opus 4.6

@techdevnotes This 1 million token context at those rates is impressive. The capabilities listed make it a strong option for complex work.

@techdevnotes Context size only matters if the model actually attends to it well. Most 1M token models fall apart past 200K. Pricing is competitive though.

From bare metal to AI workloads, SuperCloud Software Suite as part of Data Center Building Block Solutions delivers end-to-end management, unifying infrastructure control, automating deployment, and optimizing multi-tenant GPU cloud operations.

@techdevnotes Solid pricing.

@techdevnotes I still can't use it on the Grok interface. They're still asking for $30 😓

@techdevnotes grok-4-1-fast-reasoning is way cheaper

@techdevnotes @grok is paid now