⚠️ October 26 outage review: 10% off
Default and SVIP usage on October 28 — The discount applies automatically throughout October 28 (UTC+8), following the October 26 main incident window of 09:48–11:10 (UTC+8). Customers who temporarily removed IP allowlists because of 401 errors should restore their original configurations.📖 View details⚠️ Historical October 26 IP allowlist notice: restore temporarily removed configurations — Reverse proxy mechanisms affected IP allowlist validation, causing some tokens to return
401; customers who temporarily removed allowlists should restore their original configurations. The subsequent incident review and compensation details are now available.📖 View details⚠️ Initial October 26 recovery record; the subsequent incident review is now available — The main incident window was 09:48–11:10 (UTC+8), with some migration-related issues continuing after core API service resumed. We take responsibility; see the subsequent review for the incident details and compensation arrangements.📖 View details
📖 Seedance 2.5 Draft mode documentation updated — Generate a 480p draft with
draft: true, then reference its task ID to create a 1080p final video. Both stages are billed separately, and the draft ID is valid for seven days; parameters and complete examples are in the guide.📖 View details⚠️ Image quality is fluctuating on some GPT Image models — A slight quality decline on
gpt-image-2-all and gpt-image-2.5-all has been traced by our investigation to an OpenAI-side compute issue. Small Chinese text is inconsistent on gpt-image-2-vip; we suspect provider-side fluctuations, possibly involving quality tier adjustments, and continue monitoring.📖 View details⏸ Veo 3.1 Preview has been retired, and APIYI has disabled the corresponding channels — Google shut down the models on October 22; related 404 errors reflect their retirement. Omni is still planned, with a possible November launch subject to a further announcement.📖 View details
🚀 Kimi K3.1 Pro and Highspeed launch via official API forwarding — Basic text, streaming, JSON, tool round-trip, and image-understanding tests passed. Rates match official pricing; eligible top-ups with 20% bonus credit bring effective costs to about 83.3% of list price. See details for cache validation limits.📖 View details
⚠️ Railway users: rotate API keys; all users: secure your accounts — Following Railway’s environment-variable security incident, revoke and replace unrotated keys stored there and update deployments. Enable APIYI 2FA or a passkey, cap test-key spending, and allowlist stable public egress IPs for production keys.📖 View details
⏸ OpenAI retired 12 legacy model entries and five additional fine-tuned model categories on October 23. Please migrate promptly — Affected entries include specified GPT-3.5, GPT-4, GPT-4.1 Nano, GPT-4o, o-series, and GPT Image snapshots and aliases; check applications and workflows for old references, update
model, and validate behavior.📖 View details📖 New Token Management API scenario: a daily budget per key that resets the next day — There is no built-in daily quota, but a scheduled job at 00:00 (UTC+8) that sets each key’s
remain_quota back to its daily cap does it; keys don’t change, calls stop without charge once the day’s budget is used, and they work again the next day. A ready-to-run script is included.📖 View details📖 For earlier updates, visit the Live Updates Archive — browse history by month, category, or vendor.
Deep Dive
AI Radar
Telegram
Global users
Media Models
New additions