Key Info
Tencent announced Hy4 Flash, described as its best-value Hy model to date, priced at $0.07 per 1M input tokens and $0.281 per 1M output tokens, with cache hits at $0.003 per 1M tokens.
Highlights
- Claims everyday tasks run at 1/20 the cost of Hy4, with throughput of 150+ tokens per second.
- In human blind tests, it reportedly matches Claude Haiku 5.5 (high) on daily work while cutting task cost by 18x.
- Promotional offer through Nov 30: free on WorkBuddy and TokenHub API at half price; OpenRouter support coming soon.