Hacker News: mgw

New comment by mgw in "Ask HN: What are you working on? (June 2026)"

mgw — Sun, 14 Jun 2026 20:26:30 +0000

Mainly https://lumenfall.ai, which is OpenRouter for media models, with mock generations for testing and judge models for production model performance monitoring. We also have our own arena where we benchmark image, video and svg models.

As a side project I‘m building a multi agent harness that works across desktop and mobile and solves the issue of drowning in too many agent sessions for my own workflow. Hopefully I’ll open source it soon. Reach out if you’d like to beta test it. (Email in my profile)

New comment by mgw in "Ask HN: What are you working on? (May 2026)"

mgw — Sun, 10 May 2026 21:14:37 +0000

I‘m building a way to record and replay AI image and video generation API calls to any provider so that in testing, people can save money. This is part of our AI media model gateway https://lumenfall.ai

New comment by mgw in "Eden AI – European Alternative to OpenRouter"

mgw — Sun, 26 Apr 2026 12:14:23 +0000

We have direct relationships with providers and get volume discounts.

Besides that, we‘ll introduce paid plans with additional features like observability soon.

New comment by mgw in "Eden AI – European Alternative to OpenRouter"

mgw — Sun, 26 Apr 2026 11:57:08 +0000

Eden AI is a solid product and it’s good that there are alternatives to OpenRouter out there.

We don’t brand ourselves as such, but if you’re looking for a European OpenRouter alternative focusing on media models (image, video etc.), I‘ve built that with https://lumenfall.ai

New comment by mgw in "Ask HN: What Are You Working On? (April 2026)"

mgw — Sun, 12 Apr 2026 23:35:30 +0000

We‘ve built an AI image and video model gateway called https://lumenfall.ai.

Right now I‘m working on adding a „simulation“ mode, that allows anyone to get free fake responses during development, instead of pricey real generations.

New comment by mgw in "StepFun 3.5 Flash is #1 cost-effective model for OpenClaw tasks (300 battles)"

mgw — Wed, 01 Apr 2026 19:16:37 +0000

Missing from the comparison is MiMo V2 Flash (not Pro), which I think could put up a good fight against Step 3.5 Flash.

Pricing is essentially the same: MiMo V2 Flash: $0.09/M input, $0.29/M output Step 3.5 Flash: $0.10/M input, $0.30/M output

MiMo has 41 vs 38 for Step on the Artificial Analysis Intelligence Index, but it's 49 vs 52 for Step on their Agentic Index.

New comment by mgw in "Ask HN: What Are You Working On? (March 2026)"

mgw — Mon, 09 Mar 2026 13:28:02 +0000

https://lumenfall.ai

Right now we are "OpenRouter for Images", with video following this week.

Our north star is creating a broader developer platform for AI media generation that includes observability, with fine-tuned vision models as a judge to monitor production traffic.

We also have a model arena and showdown page that ranks models by task, so you can find the best model for e.g. photorealism: https://lumenfall.ai/leaderboard

Our stack is Rails for the dashboard and Cloudflare Workers (Typescript / Hono) for the engine.

Helicone Is Joining Mintlify

mgw — Tue, 03 Mar 2026 18:01:09 +0000

Article URL: https://www.helicone.ai/blog/joining-mintlify

Comments URL: https://news.ycombinator.com/item?id=47236209

Points: 5

# Comments: 1

New comment by mgw in "Ask HN: What are you working on? (February 2026)"

mgw — Mon, 09 Feb 2026 06:22:08 +0000

https://lumenfall.ai

A developer platform for AI image generation that includes observability, with fine-tuned vision models as a judge to monitor production traffic. (Still working on the last part.)

We also have a model arena and showdown page that ranks models by task, so you can find the best model for e.g. making infographics: https://lumenfall.ai/leaderboard

We just launched the MVP. Tech stack is Rails for the dashboard and Cloudflare Workers (Typescript / Hono) for the gateway.

New comment by mgw in "Show HN: Gigacode – Use OpenCode's UI with Claude Code/Codex/Amp"

mgw — Sat, 07 Feb 2026 06:36:57 +0000

This looks neat. Thank you!

How does this relate to the recent changes of not being able to use a Claude Code Max subscription outside of Claude Code anymore? I assume technically this would work with Gigacode but would still be against their ToS?

New comment by mgw in "ChatGPT Atlas"

mgw — Wed, 22 Oct 2025 09:06:51 +0000

Very interesting that you can switch on having website content summarized locally. I wonder which model they are using for that.

https://help.openai.com/en/articles/12574142-chatgpt-atlas-d...

# Enable on-device summaries: Users on macOS 26 have the option to have web content summarized on their device, so that web contents are not sent to our servers.

New comment by mgw in "I couldn't submit a PR, so I got hired and fixed it myself"

mgw — Sat, 02 Aug 2025 08:18:21 +0000

Another Google Maps request: Places I‘ve labelled used to at least sometimes (if inconsistently) show up in search. Now they never shop up, even when I type the exact name of the label.

New comment by mgw in "Infomaniak comes out in support of controversial Swiss encryption law"

mgw — Fri, 06 Jun 2025 11:33:53 +0000

A combination of good pricing, sane behavior and an offering with many TLDs right now is https://www.netim.com, based in France. Their UI is quite outdated, but it works ok. I've switched here after leaving Gandi.

If you're fine with a US-based provider, https://porkbun.com/ also has good pricing and a tech oriented mindset. They don't support many ccTLDs though.

In general, https://tld-list.com/ is the best place to research domain registrars in my opinion.

New comment by mgw in "Business books are entertainment, not strategic tools"

mgw — Sat, 10 May 2025 05:09:22 +0000

Where does the point of being disagreeable come from and what purpose does it serve in the business world, in your opinion or according to these books?

New comment by mgw in "Gemini 2.5 Pro Preview"

mgw — Tue, 06 May 2025 15:55:01 +0000

This has also been my biggest gripe with Gemini 2.5 Pro. While it is fantastic at one-shotting major new features, when wanting to make smaller iterative changes, it always does big refactors at the same time. I haven't found a way to change that behavior through changes in my prompts.

Claude 3.7 Sonnet is much more restrained and does smaller changes.

Waymo's Research on an End-to-End Multimodal Model for Autonomous Driving

mgw — Mon, 04 Nov 2024 09:39:41 +0000

Article URL: https://waymo.com/blog/2024/10/introducing-emma/

Comments URL: https://news.ycombinator.com/item?id=42039959

Points: 2

# Comments: 0

New comment by mgw in "How to Become a Registrar (2012)"

mgw — Mon, 22 Jul 2024 05:43:55 +0000

One idea I‘ve been toying with is to set up a non-profit registrar funded by donations. It would just pass through the registration fees from the registries without markup. Similar to Let‘s Encrypt but for domains. Of course it would be more complicated because the entity would handle money, but nothing that couldn‘t be solved from what I can see.

Does anyone with experience in the field have any insights on what roadblocks would be encountered?

What's Next for Kagi?

mgw — Thu, 30 May 2024 16:01:32 +0000

Article URL: https://blog.kagi.com/what-is-next-for-kagi

Comments URL: https://news.ycombinator.com/item?id=40525249

Points: 140

# Comments: 31

New comment by mgw in "Stripe is no longer a suitable payment processor"

mgw — Wed, 02 Aug 2023 06:21:31 +0000

I would recommend Adyen. It‘s a bit like a European Stripe.

https://www.adyen.com/

New comment by mgw in "Why do cloud providers keep building datacenters in Phoenix?"

mgw — Sat, 15 Jul 2023 14:56:54 +0000

The Register misquotes itself by writing that “Google's datacenters in Dallas, Texas consumed more than a quarter of the city's water supply”.

Clicking through to the quoted article [1], the figure is actually for The Dalles, Oregon, a city with a population about 80 times smaller than Dallas’.

[1] https://www.theregister.com/2022/12/19/google_datacenters_da...