Tencent launches Hy4 flash with 224B parameters and a million-token context

Hy4 flash launched on November 16 with 224B total and 12B active parameters, multimodal support and a 1M-token context. Standard API rates are RMB 0.5 input and RMB 2 output per million tokens.

Hy4 flash is available

Tencent announced Hy4 flash on November 16, 2026, targeting everyday work at lower cost. It has 224B total parameters, including 64B Engram parameters, with 12B active parameters, multimodal support and a 1M-token context.

Pricing and access

Published API rates per million tokens are RMB 0.02 for cached input, RMB 0.5 for ordinary input and RMB 2 for output.

The model is available in Tencent Cloud TokenHub, WorkBuddy, Yuanbao, WeCom, QQ, ima and Marvis. OpenRouter integration is planned. Tencent advertises free WorkBuddy use and a 50% TokenHub API discount before November 30; check product pages for quotas and applicable terms.

Where it fits

Tencent recommends deep search, document processing, presentations, simple frontend work and computer use, while positioning Hy4 for more complex software engineering. Its claimed decoding speed exceeds 150 tokens per second; TboxIO has not independently verified that result.

Open sourcing is planned, rather than confirmed downloadable weights. Evaluate output quality, completion time and total cost on your own tasks.

Related: Qwen4-Max.

Sources

Source labels distinguish official claims, third-party tests, and in-house tests.

← Back to AI News