Hacker News: theHocineSaad

Kimi K3 and GLM-5.3 are better than Gemini 3.8 Flash

theHocineSaad — Wed, 02 Sep 2026 17:11:49 +0000

Based on artificialanalysis.ai, Kimi K3 and GLM-5.3 are more intelligent than the new Gemini 3.8 Flash[1][2].

Gemini 3.8 Flash comes in eighth place with a score of 59, just after Kimi K3 and GLM-5.3, with scores of 60 for both of them.

Gemini 3.8 Flash was better in terms of Cost per task, with $0.58 spent to finish the task, $0.68 for GLM-5.3, and 0.84 for Kimi K3[3][4].

Because results may change over time, I included screenshots taken at the time of writing this post.

[1] https://artificialanalysis.ai/#intelligence [2] https://imgur.com/a/BMOJBED

[3] https://artificialanalysis.ai/#price-and-cost [4] https://imgur.com/a/55sQRUZ

Comments URL: https://news.ycombinator.com/item?id=49539315

Points: 8

# Comments: 2

New comment by theHocineSaad in "Gemini 3.8 Flash and 3.8 Flash Cyber"

theHocineSaad — Wed, 02 Sep 2026 16:08:20 +0000

As of writing this comment, Claude Opus 5 has an intelligence score of 63, not 59 (it's not the same as Gemini 3.8 Flash).

With a score of 59, Gemini 3.8 Flash is in eighth place, falling behind even Grok 4.6, Kimi k3, and GLM 5.3.

https://imgur.com/a/BMOJBED

Go 1.27 Released with Generic Methods, JSON v2, and Faster Memory Allocation

theHocineSaad — Fri, 21 Aug 2026 10:28:32 +0000

Article URL: https://linuxiac.com/go-1-27-released-with-generic-methods-json-v2-and-faster-memory-allocation/

Comments URL: https://news.ycombinator.com/item?id=49386122

Points: 1

# Comments: 0

New comment by theHocineSaad in "Show HN: RAGless – similar to RAG, but $0 LLM API costs at runtime"

theHocineSaad — Fri, 14 Aug 2026 09:57:15 +0000

This is interesting, especially if the data doesn't change much.

I burned $157,143 in tokens last month. I paid $400

theHocineSaad — Fri, 14 Aug 2026 09:51:17 +0000

Article URL: https://www.orcdev.com/blog/ai-subscriptions

Comments URL: https://news.ycombinator.com/item?id=49296597

Points: 1

# Comments: 0

The Job Board Is Lying to You

theHocineSaad — Wed, 05 Aug 2026 19:33:08 +0000

Article URL: https://nanomicon.com/blog/260730-the-job-board-is-lying-to-you

Comments URL: https://news.ycombinator.com/item?id=49187792

Points: 32

# Comments: 12

New comment by theHocineSaad in "Claude Opus 5"

theHocineSaad — Fri, 24 Jul 2026 18:36:44 +0000

Opus 5 is considered the most intelligent model[0], while it's half the price of Fable 5[1], and Anthropic is still positioning Fable 5 as the most capable model[2].

Is it because maybe Anthropic engineered Opus 5 to work well on benchmarks and didn't do the same thing to Fable 5, or is there another reason?

[0]: https://artificialanalysis.ai/#intelligence

[1]: https://platform.claude.com/docs/en/about-claude/pricing

[2]: https://platform.claude.com/docs/en/about-claude/models/over...

EU approves Saudi PIF's $55B acquisition of Electronic Arts

theHocineSaad — Thu, 23 Jul 2026 14:37:28 +0000

Article URL: https://www.globalbankingandfinance.com/saudi-pifs-55-billion-ea-deal-approved-eu-merger-rules/

Comments URL: https://news.ycombinator.com/item?id=49022360

Points: 9

# Comments: 0

Using Go for Mobile Apps

theHocineSaad — Wed, 15 Jul 2026 09:28:25 +0000

Article URL: https://www.davidsobsessions.com/p/one-year-of-gomobile/

Comments URL: https://news.ycombinator.com/item?id=48918270

Points: 55

# Comments: 25

New comment by theHocineSaad in "Building a real-time AI tutor for 5-year-olds"

theHocineSaad — Fri, 10 Jul 2026 15:55:54 +0000

AI for kids? Sounds like a bad idea

OpenAI is shutting down its AI browser Atlas

theHocineSaad — Fri, 10 Jul 2026 15:47:52 +0000

Article URL: https://techcrunch.com/2026/07/09/openai-is-shutting-down-atlas-but-its-ai-browser-ambitions-are-still-growing/

Comments URL: https://news.ycombinator.com/item?id=48861557

Points: 4

# Comments: 2

New comment by theHocineSaad in "PostHog Open Sourced"

theHocineSaad — Thu, 09 Jul 2026 16:40:57 +0000

From the repo's about section: PostHog FOSS is a read-only mirror of PostHog, with all proprietary code removed. NOTE: This repo is synced automatically from the main PostHog repo. Please raise any issues and PRs there.

New comment by theHocineSaad in "PostHog FOSS"

theHocineSaad — Thu, 09 Jul 2026 16:39:06 +0000

Open sourced ?! I believe it was open source from day one.

New comment by theHocineSaad in "Department of Commerce has lifted export controls on Claude Fable 5 and Mythos 5"

theHocineSaad — Wed, 01 Jul 2026 16:31:40 +0000

Okay, but what has changed?

New comment by theHocineSaad in "Claude Sonnet 5"

theHocineSaad — Wed, 01 Jul 2026 13:01:12 +0000

What's interesting is that Claude Sonnet 5 costs more per task ($2.29) than Opus 4.8 ($1.80), while the latter is obviously better!

It actually costs more per task than every other model. It's only cheaper than Claude Fable 5.

Source: https://artificialanalysis.ai/?cost=cost-per-task#price-and-..., as of writing this comment (the results are frequently changing)