Bohdan Vitomskyi

@alexandr_wang Can we expect a coding version of the model soon?

the muse spark API will be coming soon! we have been thrilled with the amount of excitement amongst developers who want to try muse spark inside their agentic harnesses stay tuned!

What would a language look like if it were designed for the agent, not for the human typist? That question turned into X07 (x07lang.org). What I wanted the language to do differently I did not want to bolt an "AI mode" onto a normal language. I wanted the core workflow to assume: - structured source instead of fragile text edits - explicit capability boundaries instead of hidden side effects - deterministic replay instead of flaky debugging - machine-readable diagnostics instead of prose-only error messages - local budgets instead of unbounded execution That is why X07's canonical source form is x07AST JSON, why worlds are explicit in the world model, and why the repair loop is built into the normal toolchain path. x07lang.org/blog/why-im-bu…

@Kappaemme1926 GPT-5.2(xhigh)

What’s your current coding setup? - Opus 4.6 - GPT-5.4 - Codex 5.3

@rezoundous $100 or $200 is not a huge price for a frontier model for a serious work, I guess

So are all $20 AI plans just previews now?

@Hesamation yep, gpt-5.2(xhigh) is much better to me for complex programming tasks now. Hard to trust Claude Opus TBH.

AMD Senior AI Director confirms Claude has been nerfed. She analyzed Claude's session logs from Janurary to March: > median thinking dropped from ~2,200 to ~600 chars > API requests went up 80x from Feb to Mar. less thinking and failed attempts meaning more retries, burning more tokens, and spending more on tokens > reads-per-edit dropped from 6.6x → 2.0x. model stops researching code before touching it. > model tried to bail out or ask "should i continue" 173 times in 17 days (0 times before March 8). > self-contradiction in reasoning ("oh wait, actually...") tripled. > conventions like CLAUDE.md get ignored because there's less thinking budget to cross-check edits > 5pm and 7pm PST are the worst hours, late night is significantly better. this means the thinking allocation is most likely GPU-load-sensitive.

@sama I still see the gpt-5.2(xhigh) is much better than gpt-5.4(xhigh) on difficult complex programming tasks. But gpt-5.4(xhigh) is much faster. Will we get a new model that combines the best from both of them soon?

It is very nice to see Codex getting so much love. We are launching a $100 ChatGPT Pro tier by very popular demand.

@dramaricic great job!

Only way for solo founder like me, to distribute product is to build in public. After 40 days of building in public: - I wrote 2.5k posts - Got 1710 followers - Viewed by 1.4M people - Connected to other builders - Learned a lot I would say things are going great so far. Love it. Are you building in public? 👇

@ihtesham2005 @christiangori96 you got the point

@christiangori96 who found the idea? thats the founder. if claude did it.. then who asked the question? the question asker is the founder.

Can you call yourself a founder if your entire product was built by Claude?

@karpathy I noticed how divided even experienced developers are.

Judging by my tl there is a growing gap in understanding of AI capability. The first issue I think is around recency and tier of use. I think a lot of people tried the free tier of ChatGPT somewhere last year and allowed it to inform their views on AI a little too much. This is a group of reactions laughing at various quirks of the models, hallucinations, etc. Yes I also saw the viral videos of OpenAI's Advanced Voice mode fumbling simple queries like "should I drive or walk to the carwash". The thing is that these free and old/deprecated models don't reflect the capability in the latest round of state of the art agentic models of this year, especially OpenAI Codex and Claude Code. But that brings me to the second issue. Even if people paid $200/month to use the state of the art models, a lot of the capabilities are relatively "peaky" in highly technical areas. Typical queries around search, writing, advice, etc. are *not* the domain that has made the most noticeable and dramatic strides in capability. Partly, this is due to the technical details of reinforcement learning and its use of verifiable rewards. But partly, it's also because these use cases are not sufficiently prioritized by the companies in their hillclimbing because they don't lead to as much $$$ value. The goldmines are elsewhere, and the focus comes along. So that brings me to the second group of people, who *both* 1) pay for and use the state of the art frontier agentic models (OpenAI Codex / Claude Code) and 2) do so professionally in technical domains like programming, math and research. This group of people is subject to the highest amount of "AI Psychosis" because the recent improvements in these domains as of this year have been nothing short of staggering. When you hand a computer terminal to one of these models, you can now watch them melt programming problems that you'd normally expect to take days/weeks of work. It's this second group of people that assigns a much greater gravity to the capabilities, their slope, and various cyber-related repercussions. TLDR the people in these two groups are speaking past each other. It really is simultaneously the case that OpenAI's free and I think slightly orphaned (?) "Advanced Voice Mode" will fumble the dumbest questions in your Instagram's reels and *at the same time*, OpenAI's highest-tier and paid Codex model will go off for 1 hour to coherently restructure an entire code base, or find and exploit vulnerabilities in computer systems. This part really works and has made dramatic strides because 2 properties: 1) these domains offer explicit reward functions that are verifiable meaning they are easily amenable to reinforcement learning training (e.g. unit tests passed yes or no, in contrast to writing, which is much harder to explicitly judge), but also 2) they are a lot more valuable in b2b settings, meaning that the biggest fraction of the team is focused on improving them. So here we are.

x07 - the compiled language that humans trust and coding agents love to work with for faster and safer software development: github.com/x07lang/x07, x07lang.org.