Hacker News: jklmnopqrstuvw

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

jklmnopqrstuvw — Wed, 23 Sep 2026 03:09:08 +0000

From my own testing, Claude/GPT is much faster than Deepseek.

New comment by jklmnopqrstuvw in "Gemini 3.7 Flash"

jklmnopqrstuvw — Thu, 13 Aug 2026 18:00:47 +0000

From my own testing, Gemini 3.5/3.6 Flash is better than DS v4 Flash/Pro on text ability.

New comment by jklmnopqrstuvw in "DeepSeek V4 Pro 0813"

jklmnopqrstuvw — Wed, 12 Aug 2026 20:57:07 +0000

Months ago I start making this kind of test for my own reference. At beginning I I test each model multiple times, and results always same(pass or fail). Later I test only once for new models, I trust the results.

New comment by jklmnopqrstuvw in "DeepSeek V4 Pro 0813"

jklmnopqrstuvw — Wed, 12 Aug 2026 18:05:34 +0000

I don't think so. I specifically kept this PR to test model capabilities, and I've already tested a bunch of models. Current test results show that the more advanced the model is, the easier it passes. For example, GPT-5.5 Medium fails the test(has bug), but High passed.

New comment by jklmnopqrstuvw in "DeepSeek V4 Pro 0813"

jklmnopqrstuvw — Wed, 12 Aug 2026 17:34:49 +0000

Tested both DS v4 pro 0813 and Grok 4.6 (all from openrouter) on Codex cli. Worked on a same new feature development on my project.

Deepseek 4 pro: Worked for 12m 02s - cost $0.12 - has bug.

Grok 4.6: Worked for 3m 18s - cost $ 1.41 - no bug.

New comment by jklmnopqrstuvw in "Grok 4.6"

jklmnopqrstuvw — Wed, 12 Aug 2026 17:25:34 +0000

139 points in 50mins, why this news not in front page? got many downvotes?

New comment by jklmnopqrstuvw in "Claude Sonnet 5"

jklmnopqrstuvw — Tue, 30 Jun 2026 18:19:12 +0000

From my own experience, GLM-5.2 generally cost more tokens and much more slow.

New comment by jklmnopqrstuvw in "GLM 5.2 beats Claude in our benchmarks"

jklmnopqrstuvw — Mon, 29 Jun 2026 08:10:37 +0000

>Most larger orgs have to use API pricing AFAIK.

There are Business and Enterprise plans, both have discounting.

New comment by jklmnopqrstuvw in "GLM 5.2 beats Claude in our benchmarks"

jklmnopqrstuvw — Mon, 29 Jun 2026 01:30:47 +0000

> A typical session for me with GPT is usually over a hundred dollars.

I don't think a $100 session is "typical". I use GPT for months. $20/m plus plan is enough for my daily work.

Show HN: ThreadPage – Discord Alternative, but Forums

jklmnopqrstuvw — Mon, 15 Jun 2026 14:02:22 +0000

Article URL: https://threadpage.com

Comments URL: https://news.ycombinator.com/item?id=48541433

Points: 2

# Comments: 0

OpenAI's June 2026 Report on Malicious Uses of AI [pdf]

jklmnopqrstuvw — Thu, 11 Jun 2026 21:00:17 +0000

Article URL: https://cdn.openai.com/pdf/96b559fa-c165-4575-805d-e636909e2f78/June-2026-Threat-Report.pdf

Comments URL: https://news.ycombinator.com/item?id=48496332

Points: 20

# Comments: 4

New comment by jklmnopqrstuvw in "SQLite is all you need for durable workflows"

jklmnopqrstuvw — Sat, 30 May 2026 10:21:27 +0000

even with wal, postgresql can easily reach 130k tps in pipeline mode.

DeepSeek seems to be leaking random user chat history

jklmnopqrstuvw — Mon, 25 May 2026 19:15:27 +0000

Article URL: https://breakingvibe.dev/news/deepseek-leaking-user-chat-history

Comments URL: https://news.ycombinator.com/item?id=48270523

Points: 2

# Comments: 0

New comment by jklmnopqrstuvw in "Google I/O"

jklmnopqrstuvw — Tue, 19 May 2026 17:20:39 +0000

Gemini Omni

New comment by jklmnopqrstuvw in "Ubuntu 26.04"

jklmnopqrstuvw — Fri, 24 Apr 2026 06:37:43 +0000

Ubuntu 26 + KDE Plasma 6.6 perfectly handles high-DPI scaling for me. I was originally planning to buy a Mac, but luckily I saw the news about Ubuntu 26 being released a few days ago.

New comment by jklmnopqrstuvw in "404 Deno CEO not found"

jklmnopqrstuvw — Sat, 21 Mar 2026 19:49:30 +0000

I think the only real perk of Bun/Deno over Node is that they can run .ts directly and have better ES compatibility. I don’t need Deno’s permission system, but it’s on by default, so I went with Bun.