Hy3 preview
Description
Hunyuan Hy3 preview is designed for Agent workloads, adopting 295B/21B activated MoE architecture. Within the same model, it provides three modes: no_think (ultra-fast response), think_low (fast thinking), think_high (deep reasoning), adapting to different latency and depth needs from high-frequency interaction to complex engineering tasks. Approaching current best levels on code benchmarks like SWE-bench Verified, 256K context supports cross-file code refactoring and long document analysis. Suitable for developers who need reliable task completion while being sensitive to inference costs.
Specifications
- Type
- Chat
- Vendor
- Tencent Cloud
- Model ID
-
hy3-preview - Context Window
- 256K
- Max Input
- 192K
- Max Output
- 128K
- Requests / Minute
- 60
- Tokens / Minute
- 1000000
Pricing
Token<16k
Input + Output- Input Base Price
- $0.0012 / 1000 tokens
- Output Base Price
- $0.004 / 1000 tokens
Token<16k Cache Hitin
Input Only- Input Base Price
- $0.0004 / 1000 tokens
16k<=Token<32k
Input + Output- Input Base Price
- $0.0016 / 1000 tokens
- Output Base Price
- $0.0064 / 1000 tokens
Token>=32k
Input + Output- Input Base Price
- $0.002 / 1000 tokens
- Output Base Price
- $0.008 / 1000 tokens
16k<=Token<32k Cache Hitin
Input Only- Input Base Price
- $0.0006 / 1000 tokens
Token>=32k Cache Hitin
Input Only- Input Base Price
- $0.0008 / 1000 tokens