Hacker News: inferencecoder

New comment by inferencecoder in "Running Kimi K3 on MI355X at Better Performance per Dollar Than B300"

inferencecoder — Mon, 03 Aug 2026 03:08:03 +0000

> Not to mention no one serious is serving this on 8xB200 instead of multiple nodes: the vast majority of Moonshot's inference work is focused on PD-disaggregation

The GPU price discourse is absurd, but many are serving models on single node setups when the model fits

New comment by inferencecoder in "Running Kimi K3 on MI355X at Better Performance per Dollar Than B300"

inferencecoder — Mon, 03 Aug 2026 03:02:46 +0000

To begin with, as I pointed out separately, the price comparison that makes the headline is absurd. No one is easily getting the MI355X's at $2.5/hr. And then the eval is unreplicable and poorly defined.

New comment by inferencecoder in "Running Kimi K3 on MI355X at Better Performance per Dollar Than B300"

inferencecoder — Sun, 02 Aug 2026 10:46:58 +0000

I think it's fair to expect an extensive review of an article before publishing. Not everything has a set of serious flaws.

New comment by inferencecoder in "Running Kimi K3 on MI355X at Better Performance per Dollar Than B300"

inferencecoder — Sun, 02 Aug 2026 09:19:29 +0000

https://chatgpt.com/share/6a6f09ff-2830-83ea-9578-de3016cfae...

New comment by inferencecoder in "Running Kimi K3 on MI355X at Better Performance per Dollar Than B300"

inferencecoder — Sun, 02 Aug 2026 06:54:26 +0000

Wafer is making themselves synonymous with slop in the inference space. Exaggerated unfair comparisons in all their results, twitter hype posts with alarm emojis etc.

> $2.50/GPU-hr for the MI355X, $6.00 for the B300, and $4.25 for the B200.

This is not an accurate price comparison for real terms.

New comment by inferencecoder in "Running Kimi K3 on MI355X at Better Performance per Dollar Than B300"

inferencecoder — Sun, 02 Aug 2026 06:52:21 +0000

There is barely any effort, a simple GPT5.6 sol pro query rips the post apart.