• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
  • HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI

    Theo Reports Low Fable Usage Despite Heavy Testing

    Replies to users complaining about rapid Fable 5.1 limit depletion on the Max plan.

    MB
    DR
    LB
    26 Sources, 29d ago, first seen 29d ago

    TLDR

    Multiple users posted complaints that Fable 5.1 burns through session and weekly limits far faster than prior versions, with some exhausting weekly allowances in one or two sessions. Theo replied directly that he has tested heavily yet remains at 14 percent of one weekly limit, with his other accounts untouched. The exchange centers on rate-limit behavior for the Max 20x plan, where earlier posts described five-hour limits vanishing in roughly twenty minutes of intensive use.

    Combined views

    894.1K

    26 Sources, first seen 29d ago

    Combined views

    894.1K

    26 Sources, first seen 29d ago

    7.7K likes
    7.7K likes
    1K comments
    482 saves
    224 reposts
    1K comments
    482 saves
    224 reposts

    Sentiment

    Positive28.2%71.8%Negative

    Summary

    Sentiment

    Positive28.2%71.8%Negative

    Replies criticized Fable 5.1 for rapidly exhausting tokens and rate limits on high-tier plans like the $100 tier, making it unusable or overpriced, while some accounts claimed it ran efficiently overnight or preferred cheaper alternatives.

    Based on 86 sentiment-bearing replies from 85 accounts across 7 conversations.

    Summary

    Replies criticized Fable 5.1 for rapidly exhausting tokens and rate limits on high-tier plans like the $100 tier, making it unusable or overpriced, while some accounts claimed it ran efficiently overnight or preferred cheaper alternatives.

    Based on 86 sentiment-bearing replies from 85 accounts across 7 conversations.

    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    26 Sources

    @mattshumer_I'm about to run through my second Anthropic sub. I've been running three Gauntlet Loops at once on Fable 5.1, but that might not be sustainable if I want to stay solvent. I'm going to focus all of my compute on one really ambitious project 👀👀
    @theoSomehow missed that they did a reset. About to be a very fun day :)
    @kimmonismusLiterally unusable. The rate limits are absurd. Oh, and by the way, Fable’s automatic continuation is bugged and doesn’t even work. I honestly don’t know why I still bother using Claude at this point. 5.6 is simply better overall anyway. Give me GPT-Astra and im fine. its so frustrating. seriously. oh, and btw. For subscription users, Anthropic has not announced lower prices or higher usage limits regarding Fable 5.1s efficency gains; the savings explicitly apply “wherever usage is billed by token,” so greater efficiency within Pro or Max subscriptions possible not gonna happen.
    @MLStreetTalkFable 5.1 is mind-blowingly good, but the limits on the Max 20x plan are a joke. - You can easily burn through the 5 hour limit in about 20 minutes - The "weekly" limit seems to be about 2x 5 hour limits' worth. Seriously?! So basically you can burn a week's worth of tokens in about 1 hour of heavy usage. How can we embark on any work knowing we are perpetually about to run out of tokens and be left STRANDED in a bad state ... to THEN get relegated to the bleak Opus5 model, scraping the barrel with the few tokens left over. To make matters worse, there is a "50% weekly limit boost for Claude Code" until September 13. Yes - it's about to get EVEN WORSE !!! Praying for @OpenAI to release a model roughly on a par with Fable that is actually economical to use. I am losing the will to live over here. Can we just have a decent, economical model already?
    @roydanroy@MLStreetTalk Use the API.
    @TuhinChakr🤣🤣🤣🤣
    @jasondeanleeFable 5.1 after about 10 minutes on the $100 plan
    @petergyangDamn I'm running out of Fable limit already, anyone in the same boat?
    @DanielleFongi have not been hitting limits, because i run with stripped system prompt, so it's fairly efficient with tokens. i think things like websearch and some tool calls and checking can be extremely expensive, with fable
    @peterjliuI can confirm that this is normal amount of accounts with Fable 5. With Fable 5.1, I think you need more like 3-4x since it's better at orchestrating sub-agents and tackling longer tasks.

    26 Sources

    @mattshumer_I'm about to run through my second Anthropic sub. I've been running three Gauntlet Loops at once on Fable 5.1, but that might not be sustainable if I want to stay solvent. I'm going to focus all of my compute on one really ambitious project 👀👀
    @theoSomehow missed that they did a reset. About to be a very fun day :)
    @kimmonismusLiterally unusable. The rate limits are absurd. Oh, and by the way, Fable’s automatic continuation is bugged and doesn’t even work. I honestly don’t know why I still bother using Claude at this point. 5.6 is simply better overall anyway. Give me GPT-Astra and im fine. its so frustrating. seriously. oh, and btw. For subscription users, Anthropic has not announced lower prices or higher usage limits regarding Fable 5.1s efficency gains; the savings explicitly apply “wherever usage is billed by token,” so greater efficiency within Pro or Max subscriptions possible not gonna happen.
    @MLStreetTalkFable 5.1 is mind-blowingly good, but the limits on the Max 20x plan are a joke. - You can easily burn through the 5 hour limit in about 20 minutes - The "weekly" limit seems to be about 2x 5 hour limits' worth. Seriously?! So basically you can burn a week's worth of tokens in about 1 hour of heavy usage. How can we embark on any work knowing we are perpetually about to run out of tokens and be left STRANDED in a bad state ... to THEN get relegated to the bleak Opus5 model, scraping the barrel with the few tokens left over. To make matters worse, there is a "50% weekly limit boost for Claude Code" until September 13. Yes - it's about to get EVEN WORSE !!! Praying for @OpenAI to release a model roughly on a par with Fable that is actually economical to use. I am losing the will to live over here. Can we just have a decent, economical model already?
    @roydanroy@MLStreetTalk Use the API.
    @TuhinChakr🤣🤣🤣🤣
    @jasondeanleeFable 5.1 after about 10 minutes on the $100 plan
    @petergyangDamn I'm running out of Fable limit already, anyone in the same boat?
    @DanielleFongi have not been hitting limits, because i run with stripped system prompt, so it's fairly efficient with tokens. i think things like websearch and some tool calls and checking can be extremely expensive, with fable
    @peterjliuI can confirm that this is normal amount of accounts with Fable 5. With Fable 5.1, I think you need more like 3-4x since it's better at orchestrating sub-agents and tackling longer tasks.