Hacker News: PhilippGille

New comment by PhilippGille in "MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per second"

PhilippGille — Tue, 09 Jun 2026 06:36:36 +0000

The interesting bits on how they achieved it:

> On the model side, we applied FP4 quantization

> introduced DFlash, an efficient speculative decoding method based on block-level masked parallel prediction

> On the system side, TileRT perfectly adapts to the dynamic characteristics of these algorithms

> 1000+ tokens/s output [...] using just a single standard 8-GPU commodity node

New comment by PhilippGille in "MiniMax M3: The First Open-Weights Model to Combine Three Frontier Capabilities"

PhilippGille — Mon, 01 Jun 2026 07:29:00 +0000

The blog post has more info: https://www.minimax.io/blog/minimax-m3

New comment by PhilippGille in "Step 3.7 Flash"

PhilippGille — Fri, 29 May 2026 22:11:23 +0000

Do you mean MiMo V2 Flash? V2.5 doesn't have a Flash version.

New comment by PhilippGille in "Quack: The DuckDB Client-Server Protocol"

PhilippGille — Tue, 12 May 2026 21:31:55 +0000

It's in the article:

> HTTP also allows the DuckDB-Wasm distribution to speak Quack natively! So DuckDB running in a browser can e.g., directly connect to a DuckDB instance running in an EC2 server using Quack.

New comment by PhilippGille in "Using Claude Code: The unreasonable effectiveness of HTML"

PhilippGille — Sat, 09 May 2026 09:32:07 +0000

Both the original Markdown spec [1] as well as CommonMark [2] clearly specify support for inline HTML. With that you can kind of get the best of both words depending on your use case.

For the most parts you just write the regular Markdown headers and paragraphs, embed images, insert tables etc without the need for any HTML tags, making it readable in source form. And if you want to embed an SVG file for example, which the author of the article mentions as one use case, you just embed the SVG directly, and people can render the Markdown in their favorite viewer.

Let's say you're viewing a raw Markdown file in VS Code. You come onto an HTML tag, so you hit Cmd+Shift+V to open the preview and that's it.

Of course for full-fledged web pages with interactive buttons and fully customized styling and all of that, which the author shows in some examples, this is not feasible. But you can get very far when you have mostly text/images/tables and just want to add some extras here and there.

[1] https://daringfireball.net/projects/markdown/syntax#html

[2] https://spec.commonmark.org/0.31.2/#html-blocks

La Suite Docs v5.0.0 released

PhilippGille — Sat, 09 May 2026 07:42:57 +0000

Article URL: https://github.com/suitenumerique/docs/releases/v5.0.0

Comments URL: https://news.ycombinator.com/item?id=48072833

Points: 4

# Comments: 0

New comment by PhilippGille in "DeepSeek 4 Flash local inference engine for Metal"

PhilippGille — Thu, 07 May 2026 18:24:54 +0000

On max it uses more than twice as many tokens as on high when running the ArtificialAnalysis benchmark suite, and then it's indeed the model with the highest token usage (among the current top tier models). See the "Intelligence vs. Token Use" chart here:

https://artificialanalysis.ai/models?models=gpt-5-5%2Cgpt-5-...

New comment by PhilippGille in "Ask HN: Best Embedding Models?"

PhilippGille — Tue, 05 May 2026 06:04:56 +0000

Benchmarks only paint part of the picture, but it's still a decent place to start looking into recent models:

https://huggingface.co/spaces/mteb/leaderboard

New comment by PhilippGille in "DeepSeek v4"

PhilippGille — Fri, 24 Apr 2026 14:26:17 +0000

When you say "Gemini", which exact model do you mean? You know there are several and they vary a lot in how capable they are? Pro 3.1 Preview, 2.5 Pro (their latest non-preview pro model), Flash 3 Preview, ...

Same with GPT-5: Latest 5.5, prior 5.4, or actually the original 5 (.0)?

You can't talk about model performance without specifying the exact model.

New comment by PhilippGille in "High-Level Rust: Getting 80% of the Benefits with 20% of the Pain"

PhilippGille — Sun, 12 Apr 2026 08:30:44 +0000

> C# [...] only really works properly in Windows

What do you mean with this? Maybe you are thinking of the old ".NET Framework" runtime, which only runs on Windows? Nowadays there is ".NET Core" which runs on macOS and Linux as well.

New comment by PhilippGille in "I run multiple $10K MRR companies on a $20/month tech stack"

PhilippGille — Sun, 12 Apr 2026 07:49:57 +0000

He specifically mentions that he is using GitHub Copilot because of how Microsoft bills per request instead of token.

New comment by PhilippGille in "Old laptops in a colo as low cost servers"

PhilippGille — Fri, 10 Apr 2026 06:21:30 +0000

> it is possible with some software to have everything massively cached, with the cloud doing that, with the origin server in my basement, only accessible from the allowed cache arrangement

Do you mean a setup like:

    client -> cloud(HAProxy+Varnish) -WireGuard-> basement(backend)

Or something else?

New comment by PhilippGille in "A truck driver spent 20 years making a scale model of every building in NYC"

PhilippGille — Tue, 07 Apr 2026 21:28:21 +0000

If you are interested in scale models of New York, there's a 1:1 scale model in Minecraft: https://youtu.be/ZouSJWXFBPk

New comment by PhilippGille in "Docker Offload"

PhilippGille — Sun, 05 Apr 2026 18:50:08 +0000

The article tries to sell it to people who can't run Docker locally (e.g. locked down permissions in enterprise environments, slow old laptop), but hasn't it already been possible to use remote Docker engines?

So the news is that they're offering to host those remotes now, right?

New comment by PhilippGille in "Codex pricing to align with API token usage, instead of per-message"

PhilippGille — Sun, 05 Apr 2026 16:55:19 +0000

Is this not just about extra credit? So what's included in the subscription doesn't change - just extra credits are now token based instead of message based? (For Plus/Pro)

New comment by PhilippGille in "Qwen3.6-Plus: Towards real world agents"

PhilippGille — Thu, 02 Apr 2026 15:19:26 +0000

The OpenRouter usage stats indicate the opposite: https://openrouter.ai/rankings?view=month

New comment by PhilippGille in "43 hours battery life: Dell XPS 14 2026 lasts almost 3x longer vs MacBook Air 15"

PhilippGille — Thu, 02 Apr 2026 06:29:31 +0000

This is their MBP 14" M5 Max review, with a "Battery life" section and their standard web browsing test: https://www.notebookcheck.net/M5-Max-with-inconsistent-perfo...

15h 10min

New comment by PhilippGille in "I traced my traffic through a home Tailscale exit node"

PhilippGille — Wed, 01 Apr 2026 06:24:14 +0000

The article does list what Tailscale adds on top of WireGuard:

> WireGuard by itself is mostly the data plane. Tailscale adds the control plane on top: identity/SSO, peer discovery, NAT traversal coordination, ACL distribution, route distribution (including exit node default routes), MagicDNS, and fast device revocation.

New comment by PhilippGille in "purl: a curl-esque CLI for making HTTP requests that require payment"

PhilippGille — Wed, 25 Mar 2026 16:35:23 +0000

In the cryptocurrency world this has existed many years already. For example with the Lightning network on top of Bitcoin it has always been easy to let the server generate an invoice, which the client pays and then the client sends another request including cryptographic proof of the payment. On layer 2 it was always cheap and fast.

For example I created this Go middleware at the time: https://github.com/philippgille/ln-paywall#how-it-works (currently defunct)

Similar projects implemented that into standalone API gateways.

All using status code 402 already.

Here Stripe's example is using USDC, so still crypto BTW.

New comment by PhilippGille in "ARM to make processors for first time in their history"

PhilippGille — Tue, 24 Mar 2026 22:52:58 +0000

83 points, 24 comments: https://news.ycombinator.com/item?id=47506641