Ask HN: Google Colab alternatives for large lang models?

I just want to run some LLM experiments on GPUs. Since these are 10-20GB models, I need something like an A100 40GB GPU at a minimum.<p>But I&#x27;m finding that most of the paid Google Colab options don&#x27;t work. Instead of working on the LLM experiments, I&#x27;m wasting considerable time on infrastructure.<p>First, I purchased pay-as-you-go credits[1]. But about 80% of the time, I get a &quot;Selected GPU unavailable&quot; for A100.<p>Next, I tried the custom GCE VM option. But the Colab deployment marketplace app[2] keeps hitting me with quota denials for A2-CPUs and A100-GPUs.<p>Requesting quota increases in GCP is equally frustrating. They auto-approved 12 A2-CPUs for a zone but denied one A100-GPU for that zone!<p>There are some 50 or so zones and apparently I have to individually request quota increases each of them but they get denied.<p>-----<p>QUESTION 1:<p>How are you people testing LLMs like GPT-NeoX, LLaMA 66B, etc?<p>------<p>QUESTION 2:<p>Is there a simple, non-time-wasting notebook alternative for Colab where I can get A100 or better GPUs easily?<p>Is there some arrangement of using Colab with local runtime which somehow proxies to a better GPU service provider?<p>Kaggle NB is also frustrating in other ways. Their GPUs don&#x27;t seem to work with these models. Also difficult to connect to GDrive, difficult to transfer files in and out, etc.<p>-----<p>QUESTION 3:<p>If there&#x27;s no alternative to Colab, anybody here has some practical tips to avoid these quota denials?<p>------<p>[1]: https:&#x2F;&#x2F;colab.research.google.com&#x2F;signup&#x2F;pricing<p>[2]: https:&#x2F;&#x2F;research.google.com&#x2F;colaboratory&#x2F;marketplace.html