HN Distilled

Essential insights from Hacker News discussions

Gemini CLI

Original Article

Hacker News Discussion

Here's a breakdown of the key themes from the Hacker News discussion about the Gemini CLI, along with quotes:

Claude Code's Perceived Superiority and User Experience

A dominant theme is the positive comparison to Anthropic's Claude Code, particularly regarding its user experience and predictability. Many users appear to be switching or considering switching to it.

  • "anyone seriously engaged has switched to claude code now (probably including many devs AT openai/google/etc.)" - coolKid721
  • "If you don't think claude code is just miles ahead of other things you haven't been using it (or well)" - coolKid721
  • "I strongly prefer Claude code to aider - irrespective of the max plan." - shmoogy
  • "Generally i really prefer Claude Code's UX (CLI based, permissions, etc) - it's all generally close to right for me, but not perfect." - unshavedyak

Pricing Models and Cost Concerns (Claude vs. Aider vs. Gemini)

The discussion heavily focuses on the cost and billing models of different AI coding tools, particularly contrasting Claude Code's subscription model with Aider's API-based billing and Google's free tier. Users are concerned about unexpectedly high bills. The comfort of a fixed monthly cost is attractive.

  • "The killer feature of Claude Code is that you can just pay for Max and not worry about API billing. It lets me use it pretty much all the time without stressing over every penny or checking the billing page. Until they do that - I'm sticking with Claude." - poszlem
  • "With Aider you'll be paying per API call, but it will cost quite a bit less than the similar work if using Claude Code in API-mode." - adamcharnock
  • "Using Claude models in aider burns tokens you need to top up. With Claude Max subscription you can pay a 100 or 200 USD per month plan and use their internal tool claude code without the need to buy additional pay as you go tokens. You get a 'flatrate', the higher plan gives you more usage with less rate limiting." - therealmarv
  • "...we offer the industry’s largest allowance: 60 model requests per minute and 1,000 requests per day at no charge." - jedi3335 (quoting the Gemini CLI announcement)
  • "However i didn't use Claude Code before the Max plan because i just fret about some untrusted AI going ham on some stupid logic and burning credits." - unshavedyak
  • "This insistence by SAAS vendors upon not protecting you from financial ruin must surely be some sort of deadweight loss." - rusk

Quality and Behavior of Gemini Models (vs. Claude)

Several users express dissatisfaction with the Gemini models' verbosity, difficulty in shaping behavior, and perceived decline in quality compared to its initial promise. This is contrasted with Claude's models, which are considered more controllable and focused, even if less initially impressive.

  • "Yea, i'm not even really interested in Gemini atm because last i tried 2.5 Pro it was really difficult to shape behavior. It would be too wordy, or offer too many comments, etc - i couldn't seem to change some base behaviors, get it to focus on just one thing." - unshavedyak
  • "Sonnet/Opus via Claude Code are definitely less intelligent than my early tests of 2.5 Pro, but they're reasonable, listen, stay on task and etc." - unshavedyak
  • "The verbosity of the code is the worse. Comments often longer than the actual code, every nook and cranny of an algorithm unrolled over 100's of lines, most of which unnecessary. Feels like typical code a mediocre Java developer would produce in the early 2000's" - ur-whale
  • "I've found that Gemini 2.5 Pro is pretty good at analyzing existing code, but really bad at generating a new code." - sirn
  • "Gemini 2.0 is such a work horse for API model with great price. Gemini 2.5 flash lite same price but is not as good except math and coding (very niche use case for API key)" - pzo

Aider as a Model-Agnostic Alternative

Aider is highlighted as a valuable tool because it can be used with different API models, allowing users to choose the best model for the task at hand and avoid vendor lock-in.

  • "Aider is what you want for that." - jvanderbot (in response to needing a model-agnostic CLI)
  • "I'm happy I can switch models as I like with Aider. The top models from different companies see different things in my experiences and have their own strengths and weaknesses." - therealmarv

Speculation on Google's Motivation and Business Practices

There are some cynical comments regarding Google's motives, suggesting they are simply copying competitors and using their vast resources to dominate the market, even if it means initially offering services cheaply (or free) with the potential for price increases later.

  • "Definitely not because of Claude Code eating our lunch!" - ed_mercer
  • "And since they have essentially unlimited money they can offer a lot for free/cheaply, until all competitors die out, and then they can crank up the prices" - troupo
  • "symptomatic of Google's lack of innovation and pm's rushing to copy competitor products" - htrp

Privacy Concerns About Google's Free Offerings

A minor theme is the lack of privacy guarantees with Google's free services, with concerns raised that data may be used for training purposes with human review.

  • "Don’t know about Claude, but usually Google’s free offers have no privacy protections whatsoever — all data is kept and used for training purposes, including manual human review." - thimabi

Easter Eggs and Swag

A brief mention is made of the easter eggs that these AI coding tools sometimes contain (e.g., /corgi for Gemini, and the previous swag offer for Claude Code). This is primarily for novelty.

  • "These always contain easter eggs. I got some swag from Claude Code, and as suspected, Gemini CLI includes /corgi to activate corgi mode." - sync