The Orchestrator › Models › Tencent
Tencent: Hy4 preview
Tencent: Hy4 preview is a mixture-of-experts model from Tencent, with 49B active parameters out of 770B total. It is designed for coding agents, complex tool-use workflows, and productivity tasks that...
Specifications
| Provider | Tencent |
| Model ID | tencent/hy4-preview |
| Context window | 1.0M tokens (1,048,576) |
| Modalities | text → text |
| Tokenizer | Other |
| Router sources | OpenRouter, TrustedRouter, Vercel AI Gateway |
| Released | 2026-08-28 |
| Tracked since | 2026-11-14 |
| Popularity | #14 by usage · peak 10.6% spend share in an OpenRouter category |
| Reasoning | On by default |
Pricing (per million tokens)
| Variant | Source | API ID | Input | Output | Context |
|---|---|---|---|---|---|
| standard | OpenRouter | tencent/hy4-preview | $0.834 | $2.50 | 1.0M |
| standard | TrustedRouter | tencent/hy4-preview | $0.88 | $2.64 | 1.0M |
| standard | Vercel AI Gateway | tencent/hy4-preview | $0.834 | $2.50 | 1.0M |
Recorded price history
| Date | Input / 1M | Output / 1M | Context |
|---|---|---|---|
| 2026-11-14 | $0.834 | $2.50 | 1.0M |
| 2026-11-14 | $0.88 | $2.64 | 1.0M |
| 2026-11-14 | $0.834 | $2.50 | 1.0M |
Capabilities
Agentic tool useStructured outputsReasoning mode1M+ context windowControllable reasoningNo content storage (attested)
Trust flags come from TrustedRouter's per-provider attestations.
Supported parameters
frequency_penalty, include_reasoning, logit_bias, logprobs, max_completion_tokens, max_tokens, min_p, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p