Anecdotally this tracks what I've felt over the past month, though I haven't rigorously quantified it. I've just been burning through my quota considerably more quickly than I was a week ago. Hitting limits I didn't before. I hit my weekly max yesterday, it resets tomorrow so I asked my admin to add $50 to my overage limit so I could bridge the gap. Burned that in an hour, I was astounded. Two weeks ago that much bought me two days of work. I asked for another $10 so I could simply have Claude dump handoff notes that could be picked up by Codex, and got through 4 of the 10 agents I'd had running before running out. When it resets tomorrow I'm setting the default back to 4.7, my strong suspicion is 4.8 was designed to lighten the load by burning more quota (not necessarily tokens) on the backend. I don't understand the mechanism but they're clearly putting the squeeze on power users. Curious what others have experienced.
I'm definitely seeing 4.8 use fewer tokens than 4.7 did to accomplish the same outcomes. And the key is, I think it takes less clock time for me to get the outcomes I want with 4.8 - it's more likely to get things right the first time.
I think this is us humans mis-attributing "it's getting more done" as "it costs more". I think the rate of getting more done scaled quite a bit faster than the token burn.
I'm with you. I haven't materially been more satisfied with the code or reasoning with 4.8 than I was with 4.7. But I'm also not vibe coding, I'm reviewing all of the output. Maybe 4.8 has been making fewer mistakes that I otherwise would have corrected on, but I was perfectly happy going through a few iterations with 4.7 to get it over the finish line. This trend just has me startled and I'm now realizing that my workflow will need to shift to open-weight models very soon. They're cranking the costs and there's no way I can get my employer to cover what's apparently become $2k a day in token use.
I have one Claude doing triple duty: helping with side gigs, supporting my corporate job when the company-provided Claude tokens run out at lightning speed, and helping me develop a startup on the side, which I’m literally just voicing into existence.
im assuming theyve changed how their kv cache works by evicting users of any cluster after an hour and likely prioritize active users and poorly maintain claude code.
Anecdotally this tracks what I've felt over the past month, though I haven't rigorously quantified it. I've just been burning through my quota considerably more quickly than I was a week ago. Hitting limits I didn't before. I hit my weekly max yesterday, it resets tomorrow so I asked my admin to add $50 to my overage limit so I could bridge the gap. Burned that in an hour, I was astounded. Two weeks ago that much bought me two days of work. I asked for another $10 so I could simply have Claude dump handoff notes that could be picked up by Codex, and got through 4 of the 10 agents I'd had running before running out. When it resets tomorrow I'm setting the default back to 4.7, my strong suspicion is 4.8 was designed to lighten the load by burning more quota (not necessarily tokens) on the backend. I don't understand the mechanism but they're clearly putting the squeeze on power users. Curious what others have experienced.
I'm definitely seeing 4.8 use fewer tokens than 4.7 did to accomplish the same outcomes. And the key is, I think it takes less clock time for me to get the outcomes I want with 4.8 - it's more likely to get things right the first time.
I think this is us humans mis-attributing "it's getting more done" as "it costs more". I think the rate of getting more done scaled quite a bit faster than the token burn.
My eyes rolled so far back I'm blind
I'm with you. I haven't materially been more satisfied with the code or reasoning with 4.8 than I was with 4.7. But I'm also not vibe coding, I'm reviewing all of the output. Maybe 4.8 has been making fewer mistakes that I otherwise would have corrected on, but I was perfectly happy going through a few iterations with 4.7 to get it over the finish line. This trend just has me startled and I'm now realizing that my workflow will need to shift to open-weight models very soon. They're cranking the costs and there's no way I can get my employer to cover what's apparently become $2k a day in token use.
I have one Claude doing triple duty: helping with side gigs, supporting my corporate job when the company-provided Claude tokens run out at lightning speed, and helping me develop a startup on the side, which I’m literally just voicing into existence.
At $260 a month, it’s a bargain.
I’d pay double without blinking.
im assuming theyve changed how their kv cache works by evicting users of any cluster after an hour and likely prioritize active users and poorly maintain claude code.
theyre googling themselves at light speed.
[dead]