Hacker News: apitman

New comment by apitman in "One month coding with GLM 5.3 Flash"

apitman — Sat, 03 Oct 2026 14:32:27 +0000

Honest question: if that's the case why do my GPUs draw their full TDP during inference?

New comment by apitman in "Livenerf: Has Opus 5.5 been nerfed yet?"

apitman — Wed, 30 Sep 2026 17:44:38 +0000

It's obvious unless you have multiple requests from different models in flight at the same time, and the sum total usage comes out to less than a single percentage.

New comment by apitman in "Livenerf: Has Opus 5.5 been nerfed yet?"

apitman — Wed, 30 Sep 2026 01:44:39 +0000

Do Anthropic quotas give you precise remaining token counts or something? I have something similar set up for tracking my ChatGPT usage but it only gives percentages remaining, which is a pretty coarse metric.

New comment by apitman in "GPT 6.1 Sol: Near-Astra intelligence for a fifth of the price"

apitman — Tue, 29 Sep 2026 20:14:25 +0000

I use OpenCode Go. I've also used the DeepSeek provider through OpenRouter a bit in the past and it seemed solid.

New comment by apitman in "GPT 6.1 Sol: Near-Astra intelligence for a fifth of the price"

apitman — Tue, 29 Sep 2026 19:14:52 +0000

Money is also money, which anyone who makes heavy use of parallel subagents will quickly learn.

New comment by apitman in "GPT 6.1 Sol: Near-Astra intelligence for a fifth of the price"

apitman — Tue, 29 Sep 2026 19:13:23 +0000

What's the most important task you would/wouldn't trust with an agent below the line?

New comment by apitman in "GPT 6.1 Sol: Near-Astra intelligence for a fifth of the price"

apitman — Tue, 29 Sep 2026 18:57:15 +0000

You have a lot of control over compaction, both directly by changing compaction settings, and indirectly by how you structure your codebase/docs so agents use less tokens.

New comment by apitman in "GPT 6.1 Sol: Near-Astra intelligence for a fifth of the price"

apitman — Tue, 29 Sep 2026 18:54:02 +0000

I've been trying to use DeepSeek V4.1 Flash more and been very impressed. My current (very rough) rule of thumb is that an Artificial Analysis score of ~40 is the crossover point for "good enough" for most of the things I need to do with coding agents.

OpenAI officially allows using ChatGPT subscriptions with 3rd party apps

apitman — Tue, 29 Sep 2026 18:34:01 +0000

Article URL: https://help.openai.com/en/articles/20001542-using-your-chatgpt-plan-in-other-apps-and-sites

Comments URL: https://news.ycombinator.com/item?id=49898221

Points: 2

# Comments: 0

New comment by apitman in "What even is an OS now?"

apitman — Sun, 27 Sep 2026 16:09:56 +0000

Neither, I'm criticizing the average quality of work produced by the last ~20 years of software "engineering"

New comment by apitman in "What even is an OS now?"

apitman — Sat, 26 Sep 2026 02:05:46 +0000

Loss of control/privacy, poor performance at high cost, bad UX.

Fortunately our industry has been getting people used to these for years.

New comment by apitman in "[dead]"

apitman — Fri, 25 Sep 2026 19:08:51 +0000

I think true ASI will be when AI can bring back n-gate

New comment by apitman in "Claude Opus 5.5"

apitman — Tue, 22 Sep 2026 19:14:00 +0000

Isn't the idea that they claim to be willing to slow down if everyone does (ie governments force everyone to), but otherwise they won't slow down because they still think they'll make the best choices with superintelligence if they get there first? That's my understanding of what all the major labs claim to believe anyway.

New comment by apitman in "GPT-6 Sol and Luna"

apitman — Tue, 22 Sep 2026 19:08:22 +0000

Your employer should reconsider. Sol high is cheaper than Terra max and smarter, when measured per task. ie even if tokens are more expensive Sol can often do a job with fewer tokens.

New comment by apitman in "GPT-6 Sol and Luna"

apitman — Tue, 22 Sep 2026 19:05:35 +0000

RIP Terra

New comment by apitman in "GPT-6 Sol and Luna"

apitman — Tue, 22 Sep 2026 19:04:20 +0000

Since I spent my morning fixing a bug in my OpenAI API proxy that completely broke prompt caching and caused my usage limits to burn like kindling, really happy to see some of their new cache tooling:

* Prompt caching dashboard: https://platform.openai.com/usage?usage_section=prompt-cachi...

* Adjust reasoning effort and tool availability without breaking cache

New comment by apitman in "GPT-6 Sol and Luna"

apitman — Tue, 22 Sep 2026 19:00:49 +0000

Similar for me. gpt-5.6-sol high has been my go-to for months. One of the reasons I'm pushing myself to try open models more is because it lends some level of guarantee I can continue to use the same tool as long as I want to. And I think we may just be getting to the point the open models are >= 5.6 Sol for coding.

New comment by apitman in "Nvidia announces native GPU programming in Rust"

apitman — Thu, 17 Sep 2026 21:43:38 +0000

And

https://www.sebastianaaltonen.com/blog/no-graphics-api

New comment by apitman in "ER visits for gambling disorders doubled after expanded online gambling market"

apitman — Wed, 16 Sep 2026 18:39:46 +0000

A few months ago I read an article that compared vibe coding to gambling[0] and I've found it a useful model.

[0]: maybe this one: https://www.fast.ai/posts/2026-01-28-dark-flow/

New comment by apitman in "JetKVM Mini"

apitman — Sun, 13 Sep 2026 19:52:53 +0000

I think I misunderstood the original problem