TTencent · Released Apr 22, 2026
Hy3 preview
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...
- Released
- Apr 22 2026
- Context window
- 262K 236K out
- Input / 1M tokens
- $0.18 from Free
- Output / 1M tokens
- $0.60 from Free
Where to run it
6 providers compared
Free to try on Tencent TokenHub, OpenCode — free endpoints are often limited-time and rate-limited. Cheapest: AIHubMix at $0.17 / $0.57 per 1M input / output tokens — 6% less than the priciest option.
| Provider | Input | Output | Cache read | Context | Max output | Uptime | vs. best |
|---|---|---|---|---|---|---|---|
| Tencent TokenHubfree · may be limited-time | Free | Free | Free | 256K | 64K | — | Free |
| OpenCodefree · may be limited-time | Free | Free | Free | 256K | 64K | — | Free |
| AIHubMix | $0.17 | $0.57 | $0.051 | 256K | 128K | — | Cheapest |
| ZenMux | $0.17 | $0.57 | $0.058 | 256K | 64K | — | +1% |
| GMICloudBF16 | $0.18 | $0.60 | $0.06 | 262K | 236K | 99.9%24h | +6% |
| Kilo Gateway | $0.18 | $0.60 | $0.06 | 262K | 236K | — | +6% |
Benchmarks
Benchmark scores
No benchmark scores yet — Epoch AI hasn't evaluated Hy3 preview. Scores appear here automatically once it does.
Specs
Details
- Model ID
tencent/hy3-preview- Developer
- Tencent
- Release date
- Apr 22, 2026 ·
- Context window
- 262,144 tokens
- Max output
- 235,929 tokens
- API price
- $0.18 / $0.60 per 1M input / output tokens
- Providers
- 6
- Input types
- text
- Reasoning
- Yes
- Tool calling
- Yes
- Open weights
- Yes
- Free tier
- No
Release date and prices from OpenRouter's catalog; providers from OpenRouter and models.dev. Buying directly from a provider can cost a little differently. Speed, latency and uptime are measured on OpenRouter only; uptime is the share of OpenRouter requests each provider served over the last day. Benchmark scores are the best result per benchmark from Epoch AI, licensed CC BY 4.0. Rank counts every model tracked here with a score.