Pyyan / Language Models / Hy4 flash

Hy4 flash

Tencent's cheapest Hy model, 224B with 12B active, at $0.07 and $0.281 with a 1M context.

currentTencent ↗Verified 2026-11-17

$0.07

per M input, $0.281 output

Specifications

Context1Msource · as of 2026-11-17
Input ($/Mtok)$0.07source · as of 2026-11-17
Output ($/Mtok)$0.281source · as of 2026-11-17
LicenceProprietary (weights promised)source · as of 2026-11-17
Max output64,000source · as of 2026-11-17 · medium confidence
Parameters224B total, 12B activesource · as of 2026-11-17
Released16 Nov 2026source · as of 2026-11-17
Model IDhy4-flashsource · as of 2026-11-17

What matters

  • Released on 16 November 2026 on Tencent Cloud TokenHub, WorkBuddy, Yuanbao, WeCom, QQ, ima and Marvis
  • 224B total parameters, including 64B Engram parameters, with 12B active per token
  • Reads text and images; supports reasoning, function calling, structured output and caching
  • Speaks the OpenAI Chat Completions, OpenAI Responses and Anthropic Messages APIs
  • List price $0.07 in and $0.281 out per million tokens, cache reads $0.003; half price on TokenHub until 30 November
  • Tencent says it plans to release the weights and that OpenRouter access is coming

Best for high-volume everyday office, document and agent work at a very low price.