FAST & FRUGAL · 薄
Tencent
5 models from Tencent, living in Silvergrass Fields.
Visit the Tencent village in 3D
| Model | Context | Input | Output | Accepts |
|---|---|---|---|---|
| Hy-MT2-1.8Btencent/hy-mt2-1.8b | 8K | $0.040 | $0.180 | text |
| Hy-MT2-30B-A3Btencent/hy-mt2-30b-a3b | 8K | $0.070 | $0.290 | text |
| Hy-MT2-7Btencent/hy-mt2-7b | 8K | $0.070 | $0.290 | text |
| Hy3tencent/hy3 | 262K | $0.130 | $0.530 | text |
| Hunyuan A13B Instructtencent/hunyuan-a13b-instruct | 131K | $0.140 | $0.570 | text |
Questions and answers
Which Tencent model has the largest context window?
Hy3 (Tencent), with 262K tokens.
Which Tencent model is cheapest per input token?
Hy-MT2-1.8B, at $0.040 per million input tokens ($0.180 per million output tokens).
Which Tencent models accept images?
None in the 2026-10-24 snapshot.
Which Tencent models support reasoning?
Hy3, Hunyuan A13B Instruct (2 of 5).
Answers are computed from the 2026-10-24 catalog snapshot.