Hy3 preview
Tencent · Open weights · Released Apr 22, 2026 · tencent/hy3-preview
Best $0.18 in / $0.60 out via GMICloud · bf16
262K contextToolsReasoningPrompt cachingOpen weights
Available from 1 providerUSD / MTOK · PRICES 23 MINUTES AGO · SPEED 6 DAYS AGO
EST. MONTHLY COSTclick a row for detail · click headers to sort
| PROVIDER | INPUT | OUTPUT | CACHE | CONTEXT | QUANT | TTFT | TPS | UPTIME |
|---|---|---|---|---|---|---|---|---|
| $0.18 | $0.60 | $0.06 | 262K | bf16 | ✕ | ✕ | 100.0% |
TTFT/TPS: median (P50) over trailing 504h · chat10k profile (~10K in / 200 out) · measured via OpenRouter routing · ✕ = probed, endpoint returned no measurable stream · methodology v1.1 · measured 6 days ago
SERVE THIS MODEL AND NOT LISTED? LIST YOUR INFERENCE →
Pricing variants · USD / MTok in / out
STANDARD
$0.18 / $0.60
per MTok in / out
CACHE READ $0.06
About
Hy3 preview is a high-efficiency Mixture-of-Experts model from Tencent designed for agentic workflows and production use. It supports configurable reasoning levels across disabled, low, and high modes, allowing it to...
Specifications
Context window262,144
Max output235,929
Modalities intext
Modalities outtext
LicenseOpen weights
ReleasedApr 22, 2026
TokenizerOther
WeightsHugging Face ↗
Related
Embed badge
[](https://modelindex.ai/models/tencent/hy3-preview)Sources
OpenRouter API23 minutes ago
AGGREGATORHugging Face23 minutes ago
COMMUNITYModelIndex benchmarks6 days ago
MEASUREDVerified 23 minutes ago · list prices, not negotiated rates.