cd /news/large-language-models/claude-vs-gpt-5-6-vs-gemini-vs-deeps… · home topics large-language-models article
[ARTICLE · art-80030] src=pub.towardsai.net ↗ pub= topic=large-language-models verified=true sentiment=↓ negative

Claude vs GPT-5.6 vs Gemini vs DeepSeek Prompt Caching: I Turned It Off and Saved 20%

A developer who turned off prompt caching on a Claude Opus 5 assistant saw input costs drop 20%, from $4.61 to $3.69 for 41 requests over 8 hours, because cache writes are now billed at a 1.25x premium with a 5-minute TTL on Anthropic and a 30-minute TTL on OpenAI's GPT-5.6, making caching a net loss when cache entries expire before reuse. The developer created a 200-line script to price caching across Claude, GPT-5.6, Gemini, and DeepSeek, finding that caching can act as a 25% surcharge rather than a 90% discount.

read1 min views1 publishedJul 30, 2026
Claude vs GPT-5.6 vs Gemini vs DeepSeek Prompt Caching: I Turned It Off and Saved 20%
Image: Pub (auto-discovered)

Member-only story

I turned prompt caching off on a Claude Opus 5 assistant last week and the input bill dropped 20% — from $4.61 to $3.69 for the same 41 requests over the same 8 hours.

That is not how caching is supposed to work. Every provider doc leads with the same number: cached reads cost 90% less than fresh reads. A 10x discount. Free money for anyone with a stable system prompt.

The catch is that since July 9, 2026, when GPT-5.6 went GA across Sol, Terra and Luna, cache writes are billed on both of the two biggest APIs. OpenAI replaced its old free implicit writes with explicit breakpoints, a 1.25x write premium and a 30-minute TTL, which lands it almost exactly where Anthropic has been all along. Prompt caching stopped being a one-way bet and became a wager: you pay a premium up front for a discount you might never collect.

If your cache entry expires before anyone reads it, you paid the premium for nothing. My assistant had 12 minutes between turns and Claude’s default TTL is 5 minutes. Every single request was a cold write at 1.25x. Caching was a 25% surcharge dressed up as a 90% discount. I wrote a 200-line zero-dependency script to price out all four providers properly. Three findings I did not expect:

── more in #large-language-models 4 stories · sorted by recency
── more on @claude opus 5 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/claude-vs-gpt-5-6-vs…] indexed:0 read:1min 2026-07-30 ·