Hacker News: WanderPanda

New comment by WanderPanda in "Claude AI recovers an 11 yrs old BTC wallet holding 400k USD"

WanderPanda — Thu, 14 May 2026 18:43:45 +0000

It highly depends on the task. For math and coding, sure. But for knowledge tasks GPT-4 is wayy better than even SOTA ~100B models. For my knowledge test cases the lines get blurry at >400B

New comment by WanderPanda in "Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model"

WanderPanda — Wed, 22 Apr 2026 18:23:52 +0000

I applaud that you recently started providing the KL divergence plots that really help understand how different quantizations compare. But how well does this correlate with closed loop performance? How difficult/expensive would it be to run the quantizations on e.g. some agentic coding benchmarks?

New comment by WanderPanda in "Framework Laptop 13 Pro"

WanderPanda — Wed, 22 Apr 2026 05:00:58 +0000

I would be really interested in a podcast with the CEO where he goes a bit into the trade-offs of backwards and forwards compatibility. I can not imagine that their planning was so immaculate that there aren't any regressions that a clean slate design could have cleaned up. Nevertheless, amazing job for putting this together it looks like a phenomenal product!

New comment by WanderPanda in "Taking on CUDA with ROCm: 'One Step After Another'"

WanderPanda — Mon, 13 Apr 2026 03:16:21 +0000

This is so true! Shows a lack of care that usually doesn’t stop at just the naming

New comment by WanderPanda in "A sufficiently detailed spec is code"

WanderPanda — Thu, 19 Mar 2026 05:57:05 +0000

They are heavily post-trained on code and math these days. I don‘t think we can infer that much about their behavior from just the pre-training dataset anymore

New comment by WanderPanda in "Ggml.ai joins Hugging Face to ensure the long-term progress of Local AI"

WanderPanda — Sat, 21 Feb 2026 03:47:18 +0000

Amazing work and people should really appreciate that the opportunity costs of your work are immense (given the hype).

On another note: I'm a bit paranoid about quantization. I know people are not good at discerning model quality at these levels of "intelligence" anymore, I don't think a vibe check really catches the nuances. How hard would it be to systematically evaluate the different quantizations? E.g. on the Aider benchmark that you used in the past?

I was recently trying Qwen 3 Coder Next and there are benchmark numbers in your article but they seem to be for the official checkpoint, not the quantized ones. But it is not even really clear (and chatbots confuse them for benchmarks of the quantized versions btw.)

I think systematic/automated benchmarks would really bring the whole effort to the next level. Basically something like the bar chart from the Dynamic Quantization 2.0 article but always updated with all kinds of recent models.

New comment by WanderPanda in "GLM-4.7-Flash"

WanderPanda — Mon, 19 Jan 2026 17:14:23 +0000

I find it hard to trust post training quantizations. Why don't they run benchmarks to see the degradation in performance? It sketches me out because it should be the easiest thing to automatically run a suite of benchmarks

New comment by WanderPanda in "Making tiny 0.1cc two stroke engine from scratch"

WanderPanda — Sat, 06 Dec 2025 08:43:17 +0000

Wait but the one you linked seems to be pneumatically driven, while the op one is an actual combustion engine, right?

New comment by WanderPanda in "Google Antigravity"

WanderPanda — Wed, 19 Nov 2025 02:31:54 +0000

Small feedback if any of the Antigravity people read here: "Fast" is not a great name for the "eager" option (vs. "Planning") because "Fast" is associated with "dumb" in LLMs (fast/flash/mini). Probably "Eager" would be a more descriptive name

New comment by WanderPanda in "European Nations Decide Against Acquiring Boeing E-7 Awacs Aircraft"

WanderPanda — Thu, 13 Nov 2025 17:41:00 +0000

SWIFT is Belgian, though?

New comment by WanderPanda in "Tesla Is Recalling Cybertrucks Again"

WanderPanda — Thu, 13 Nov 2025 17:38:40 +0000

Mechanically sure, but I still feel way safer when a Tesla (of any kind) is approaching me as a pedestrian or bicyclist than any other vehicle (except maybe Waymo) because I know they will alert the driver and brake if necessary. Any other car, especially older trucks, I'm quite afraid of, based on experience.

New comment by WanderPanda in "Composer: Building a fast frontier model with RL"

WanderPanda — Wed, 29 Oct 2025 17:38:29 +0000

Makes sense! I like that you guys are more open about it. The other labs just drop stuff from the ivory tower. I think your style matches better with engineers who are used to datasheets etc. and usually don't like poking a black box

New comment by WanderPanda in "Composer: Building a fast frontier model with RL"

WanderPanda — Wed, 29 Oct 2025 17:34:13 +0000

Damn TIL, I always used > Cursor: disable completions and forgot to turn it on again I need to try snooze then!

New comment by WanderPanda in "Composer: Building a fast frontier model with RL"

WanderPanda — Wed, 29 Oct 2025 17:30:07 +0000

Why did you stop training shy of the frontier models? From the log plot it seems like you would only need ~50% more compute to reach frontier capability

New comment by WanderPanda in "Apple will phase out Rosetta 2 in macOS 28"

WanderPanda — Wed, 29 Oct 2025 08:02:22 +0000

Until it isn't

New comment by WanderPanda in "Coral NPU: A full-stack platform for Edge AI"

WanderPanda — Tue, 28 Oct 2025 21:08:16 +0000

Did you check out the STM32N6? It apparently has an h264 encoder

New comment by WanderPanda in "rlsw – Raylib software OpenGL renderer in less than 5k LOC"

WanderPanda — Wed, 22 Oct 2025 06:30:13 +0000

Amazing:

(Mar 5 2022) TinyGL 0.4.1 is out (Changelog)

(Mar 17 2002) TinyGL 0.4 is out (Changelog)

"our plans are measured in centuries"

New comment by WanderPanda in "Tinnitus Neuromodulator"

WanderPanda — Sat, 18 Oct 2025 22:06:03 +0000

I have a strong Tinnitus on one ear after an ear surgery for 8 years now. And I usually don‘t notice it for months at a time, even though it is there all the time (thanks for reminding me :p) So it’s not as bad as it might feel in the beginning. I‘m mostly bothered by my hearing being generally impaired by it. It sits at ~9kHz but it somehow still makes it significantly harder to comprehend voices.

New comment by WanderPanda in "Figure 03, our 3rd generation humanoid robot"

WanderPanda — Thu, 09 Oct 2025 19:45:15 +0000

I think this is the frontier when it comes to "unstructured":

https://youtu.be/nmEy1_75qHk

They for sure did not anticipate that the user would backflip into their robot and knock it (and himself) out :D

New comment by WanderPanda in "SEC approves Texas Stock Exchange, first new US integrated exchange in decades"

WanderPanda — Wed, 08 Oct 2025 20:57:29 +0000

Theoretically, when the market offers me an order book and I take offers on one or the other side that should be totally fair? I think until execution/fill the information should be totally between me and the exchange and no one else, right? I get that if I send a limit order that can not be filled, that that affects the market because new information is introduced (before the trade) but in the previously described case all the information going out should be after the trade already happened, right?