‹ BackHN Continuity

Thread

How GLM built its own inference infrastructure

411 points · 285 comments · whiteros_e

  1. embedding-shape · · focus · HN ↗
    I was gonna ask how people found their coding plans, and realized, have they massively ramped up the prices? Seems the middle plan is ~$80/month now, didn't that used to be like $20/month? Cheapest plan is ~$20/month currently.

    They must have hit really hard scaling limits if the prices were hiked so much so quickly.

    1. Daviey · · focus · HN ↗
      I paid $360 annual for Max plan and currently averaging about 1BN tokens a day with their frontier GLM-5.3 model. This was clearly unsustainable for them and they've dropped this package.
      1. world2vec · · focus · HN ↗
        1 billion tokens a day?!! I've done a lot of work these past 2 weeks with GLM-5.3. Like, a lot. And I've just passed 300 million tokens in total.

        Can I ask where are you using all those tokens?

        1. Daviey · · focus · HN ↗
          I have 3-5 agent harnesses with large context windows working on different applications concurrently.
          1. embedding-shape · · focus · HN ↗
            Share the resulting code from any one of those please? I've tried so many times to find a setup that facilitates parallel work + high quality results, but it's just impossible regardless of harness or model. Leave the agents alone for too long, and the entire thing just balloons out of control, and next you know you're sitting there with half a million LOC where 80% isn't even needed.
            1. Daviey · · focus · HN ↗
              Most of them are not public, but a fun thing I did was a mario cli game - <a href="https:&#x2F;&#x2F;github.com&#x2F;Daviey&#x2F;mario&#x2F;" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;Daviey&#x2F;mario&#x2F; (or `ssh mario.baby`).

              I now exclusively use <a href="https:&#x2F;&#x2F;omp.sh&#x2F;" rel="nofollow">https:&#x2F;&#x2F;omp.sh&#x2F; as my harness:

              I set it up so it never works in the main branch so subagents etc don&#x27;t step on each others toes, and only merges back when complete: <a href="https:&#x2F;&#x2F;github.com&#x2F;Daviey&#x2F;mario&#x2F;blob&#x2F;main&#x2F;.omp&#x2F;hooks&#x2F;pre&#x2F;worktree-guard.ts" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;Daviey&#x2F;mario&#x2F;blob&#x2F;main&#x2F;.omp&#x2F;hooks&#x2F;pre&#x2F;wor...

              A good AGENTS.md is essential: <a href="https:&#x2F;&#x2F;github.com&#x2F;Daviey&#x2F;mario&#x2F;blob&#x2F;main&#x2F;AGENTS.md" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;Daviey&#x2F;mario&#x2F;blob&#x2F;main&#x2F;AGENTS.md

              I then provide specifications for what I want, making sure it is unit tested.

Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.