Hacker News: redman25

New comment by redman25 in "Qwen 3.8 27B available on Cerebras at 1500 tokens/s"

redman25 — Thu, 03 Sep 2026 20:14:14 +0000

Maybe they’re gunning for speedy non-interactive pricing? Or its a limit of the technology or a business decision?

FrontierHarness Eval benchmark. Pi is on the Pareto

redman25 — Thu, 03 Sep 2026 00:45:06 +0000

Article URL: https://runta.com/blog/introducing-frontierharness-eval/

Comments URL: https://news.ycombinator.com/item?id=49544583

Points: 3

# Comments: 0

New comment by redman25 in "US strikes $1.2B deal to pay German firm to halt offshore wind projects"

redman25 — Fri, 07 Aug 2026 13:59:15 +0000

More like acting without a sense of civic duty is stupid. Or acting without a basis in evidence is stupid.

New comment by redman25 in "Qwen 3.8"

redman25 — Sun, 19 Jul 2026 13:18:00 +0000

Deepseek is one of the worst in terms of hallucination rate according to artificial analysis' benchmark: https://artificialanalysis.ai/?omniscience=omniscience-hallu...

New comment by redman25 in "Prefer strict tables in SQLite"

redman25 — Sat, 11 Jul 2026 21:47:07 +0000

A large portion of the tests are closed source unfortunately which would make it tough to create a port.

New comment by redman25 in "DSpark: Speculative decoding accelerates LLM inference [pdf]"

redman25 — Mon, 29 Jun 2026 20:22:32 +0000

I've been preferring Mimo recently. Same price as deekseek, more reliable tool calling (subjectively), and has some nice qualities in terms of prose, etc.

I've heard others say that Deepseek tends to be smarter on specific problems but that Mimo tends to more well-rounded.

New comment by redman25 in "Claude Fable 5"

redman25 — Tue, 09 Jun 2026 21:45:44 +0000

Exactly, intelligence is limited by cost and physical constraints just as much as anything. That's the thing that seems to always be missing from the run-away singularity discussions, it's treated like a perpetual motion machine.

New comment by redman25 in "Gemma 4 12B: A unified, encoder-free multimodal model"

redman25 — Wed, 03 Jun 2026 16:38:46 +0000

IDK this model release is a bit disappointing considering the community has been chomping at the bit for the 124ba4b model. There was some leaked info about it but people suspect it was not released because it was too close to gemini flash in performance.

New comment by redman25 in "Big tech's anti-labor playbook has come for Wikipedia"

redman25 — Wed, 27 May 2026 03:02:57 +0000

I too prefer my misinformation delivered with maximum confidence and no accountability.

New comment by redman25 in "A few words on DS4"

redman25 — Fri, 15 May 2026 13:46:57 +0000

What prompt had you given it?

New comment by redman25 in "MacBook Neo Deep Dive: Benchmarks, Wafer Economics, and the 8GB Gamble"

redman25 — Thu, 14 May 2026 00:06:26 +0000

I'm not OP but I work outside and use light mode. Macs are generally fairly bright as long as you aren't in direct sunlight. Solarized light mode for the win though.

New comment by redman25 in "Accelerating Gemma 4: faster inference with multi-token prediction drafters"

redman25 — Tue, 05 May 2026 20:30:02 +0000

It’s a heck of a lot faster too.

New comment by redman25 in "Utah to hold websites liable for users who mask their location with VPNs"

redman25 — Sun, 03 May 2026 16:06:14 +0000

Doesn’t ham radio not allow transmissions to be encrypted by law? That rules out most of the internet.

New comment by redman25 in "Qwen3.6-35B-A3B: Agentic coding power, now open to all"

redman25 — Fri, 17 Apr 2026 09:48:49 +0000

A strix halo machine or MAC will run at less than 20watts idle. You could leave it running.

New comment by redman25 in "Artemis II safely splashes down"

redman25 — Sat, 11 Apr 2026 10:38:49 +0000

Is 10 days enough to make walking difficult?

New comment by redman25 in "Nanocode: The best Claude Code that $200 can buy in pure JAX on TPUs"

redman25 — Sun, 05 Apr 2026 19:36:17 +0000

Not to be confused with nanocoder, the agentic coding harness.

https://github.com/Nano-Collective/nanocoder

New comment by redman25 in "Google releases Gemma 4 open models"

redman25 — Thu, 02 Apr 2026 21:38:53 +0000

Exactly, compare MoE with MoE and dense with dense otherwise it's apples and oranges.

New comment by redman25 in "Google releases Gemma 4 open models"

redman25 — Thu, 02 Apr 2026 21:31:46 +0000

200a10b please, 200a3b is too little active to have good intelligence IMO and 10b is still reasonably fast.

Empiricism and Rationalism in Software Testing

redman25 — Sat, 28 Mar 2026 10:36:31 +0000

Article URL: http://joshvoigts.com/articles/empiricism-and-rationalism-in-software-testing/

Comments URL: https://news.ycombinator.com/item?id=47553339

Points: 2

# Comments: 0

New comment by redman25 in "Every layer of review makes you 10x slower"

redman25 — Tue, 17 Mar 2026 14:13:20 +0000

It's why you always have a rollback plan. Every `up` needs to a `down`.