Early warning for the AI APIs you build on
Outages, price changes, deprecations and new models — in one feed.
2outages today
21retirements ≤ 7 days
392price changes this week
Next retirement: in 3 days (2026-10-20) — 5 models, e.g. Google: Gemini 2.5 Pro (batch)
ProviderAll 706Z.ai 97DeepSeek 88~deepseek 86OpenAI 83~Z AI 78Azure OpenAI 36Fireworks 34Anthropic 31Cohere 30Perplexity 30Groq 30Google 26~moonshotai 11Moonshot AI 7Tencent 6Qwen 5DeepSeek 4Meta Llama 4Poolside 4ByteDance Seed 3Reflection 3NVIDIA 2Moondream 2Xiaomi 2Dots Studio 1Meta 1Elorian 1Xingchen Agi 1
Today
New model: Qwen: Qwen3.7 Plus (batch)
qwen/qwen3.7-plus:batch is now available on OpenRouter. Context 262144 tokens. Pricing $0.23 in / $0.92 out per 1M tokens.
Yesterday
Qwen: Qwen3 Next 80B A3B Instruct: prompt price down (-10%)
prompt tokens changed from $0.1 to $0.09 per 1M tokens (via OpenRouter).
Thu 15 Oct
Qwen: Qwen3.6 27B: completion price down (-37%)
completion tokens changed from $3.2 to $2 per 1M tokens (via OpenRouter).
Show all 2 ▾
Qwen: Qwen3.5-9B is scheduled to retire on 2026-10-21
OpenRouter lists an expiration date of 2026-10-21 for qwen/qwen3.5-9b. Plan a migration.