Hope this is true

@techdevnotes Not quite on programming, but it will excel in other areas. Anthropic has done something special with coding. It was a helpful motivator that they cut us off @xAI and not good for their karma.

@techdevnotes Grok 4.20 be like

@techdevnotes Yeah.

@techdevnotes Looking at the architecture diagram in the second slide, the way they are handling state management...

@techdevnotes They needed Anthropic to fix Grok. Let that sink in. You don't get to dream of the student surpassing the teacher when the student couldn't even do their homework without cheating.

@techdevnotes He said it's not going to beat in coding but in general it might.

@techdevnotes Nope. I have it.

@techdevnotes 5.20 too?

@techdevnotes I think it drops today

@techdevnotes Not quite on programming, but it will excel in other areas. Anthropic has done something special with coding. It was a helpful motivator that they cut us off @xAI and not good for their karma.

@techdevnotes 👍💯

@techdevnotes Explain this so an LPN who works at a Skilled Rehabilitation Center would understand

@techdevnotes Do it

@techdevnotes Benchmarks don’t change the world. Shipping does.

@techdevnotes maybe, but coding scores wont save a product without users.

@techdevnotes Man achievement get: Talk about Grok enough to summon the Musk. Congrats!

@techdevnotes are they using code world model to increase the sample efficiency in RL ?

@techdevnotes Grok 4.20 will be a high performing strain of LLM.

@techdevnotes 🎯🎯

@techdevnotes @grok are these benchmarks accurate? And secondly does Gemini 3 Flash exceed Sonnet 4.5 in coding benchmarks?

@techdevnotes Just remember @Grok is now "military grade". 🤣🤣

@techdevnotes Arn't we at the point where they are all good enough for day to day, so you can just stick with one and wait for updates? (unless you are deep in the coding world)