Tencent
Hy3
tencent/hy3
Input
$0.132
per 1M tokens
Output
$0.528
per 1M tokens
Cached input
$0.033
per 1M tokens
Hy3 is a 295B-parameter Mixture-of-Experts model from Tencent (21B active, 192 experts with top-8 routing) built for reasoning, agentic workflows, and real-world production use. It supports a configurable reasoning effort:...
- Released
- Context window
- 262K tokens
- Max output
- 128K tokens
- Input
- text
- Output
- text
- Reasoning
- Yes
Price history
- input $0.083 → $0.132 +60% · output $0.33 → $0.528 +60%
- input $0.132 → $0.083 -37% · output $0.528 → $0.33 -37%
- input $0.083 → $0.132 +60% · output $0.33 → $0.528 +60%
- input $0.132 → $0.083 -37% · output $0.528 → $0.33 -37%
- input $0.083 → $0.132 +60% · output $0.33 → $0.528 +60%
- input $0.132 → $0.083 -37% · output $0.528 → $0.33 -37%
- input $0.083 → $0.132 +60% · output $0.33 → $0.528 +60%
- input $0.132 → $0.083 -37% · output $0.528 → $0.33 -37%
- input $0.083 → $0.132 +60% · output $0.33 → $0.528 +60%
Providers on OpenRouter
5 endpoints serve this model. Uptime is OpenRouter's share of successful requests; prices per 1M tokens.
| Provider | Input | Output | Context window | Uptime 30 min | Uptime 24 h |
|---|---|---|---|---|---|
| Tencentfp8 | $0.132 | $0.528 | 262K | 100% | 100% |
| GMICloudbf16 | $0.14 | $0.58 | 262K | 100% | 100% |
| Novita | $0.14 | $0.58 | 262K | 100% | 100% |
| Phala | $0.15 | $0.64 | 262K | 100% | 100% |
| AtlasCloudfp8 | $0.20 | $0.80 | 262K | 100% | 100% |