Home › VastLLM Catalog › Hy4 Preview

Hy4 Preview

by Tencent Hunyuan · Language Models

Context 1049K tokensReasoning / thinking

Overview

Tencent Hunyuan Hy4 Preview is an open-source hybrid-reasoning MoE model for coding, agentic tasks, tool use, and long-context workloads.

Context window1049K tokens
Max output131K tokens
Input → Outputtext → text
CapabilitiesReasoning / thinking

Pricing

$0.644 in · $1.9313 out · $0.0324 cached / 1M

UsageRate (USD)Unit
Input tokens$0.644per 1M tokens
Cached input (cache hit)$0.0324per 1M tokens
Output tokens$1.9313per 1M tokens

Official model pricing · currency: USD · pre-deduction settled by actual usage

Call it in 30 seconds

OpenAI-compatible · base URL https://api.tunerouter.com/v1/

POST/chat/completions
curl https://api.tunerouter.com/v1/chat/completions \
  -H "Authorization: Bearer $TUNEROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "hy4-preview", "messages": [{"role": "user", "content": "Hello!"}]}'

Model ID: hy4-preview · works with OpenAI SDKs — just point base_url at api.tunerouter.com

More from Tencent Hunyuan

Tencent Hy3chat

$0.1019 in · $0.4077 out · $0.0255 cached / 1M

← Back to full catalog (123 models)