Unverified50% confidenceFactExact time
The UplinkSwitch aggregation platform extracted 66 chat models from 11 developers with official list prices from the real-time catalog of its own API gateway; all 66 models are below official list prices on both input and output sides, with the largest price difference reaching 95%.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/17/2026
First Seen
Valid until: 10/31/2026
Sources
Related Entities
Related Claims
UnverifiedBy billing type, the input token distribution is: fixed-subscription logins (Claude, xAI, Cursor, ChatGPT, Google Antigravity) carry about 78%; a 10-dollar-per-month coding plan carries about 11%; free model tiers (OpenRouter free and preview SKUs, NVIDIA free tier, local free routes) carry about 10%; Google API keys carry about 2%.60% similarUnverifiedIn live listing tests, pricehunt found that 35 of 59 discount cards had price parsers that mistakenly used the strikethrough price; when AliExpress showed free shipping the extension added shipping costs; the image CDN server returned 403 pages for repeated requests and these were cached by the browser for a year, and these issues were fixed in versions v1.5 and v1.6 by analyzing developer logs.59% similarUnverified在1.78百万token的长会话实测中，因为97%的token被缓存，发送token最多的配置反而最便宜59% similarUnverifiedCompared to a scheme with seven independent trajectories, shared prefix reuse reduces student-generated token costs by approximately 31% on WebShop.58% similarUnverifiedThe prompt caching feature introduced by Anthropic and OpenAI operates on the server side, offering a 50–90% price discount for content with identical long prefixes.57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1057007API
curl https://kongchang.com/api/v1/knowledge/claims/1057007MCP
get_claim(id=1057007)