Quality
Highest overall scores
- 1
Tencent: Hy3 preview41.2 - 2Tencent: Hunyuan A13B Instruct0
- 3Tencent: Hy30
Tencent model benchmarks.
Compare Tencent models across answer quality, coding, complex tasks, speed, price, and context size.
Response speed
Fastest output in this sample
- 1Tencent: Hy351 t/s
- 2Tencent: Hy3 preview27 t/s
- 3Tencent: Hunyuan A13B Instruct2 t/s
Input price
Lowest price per 1M tokens
- 1Tencent: Hy3 preview$0.06/1M
- 2Tencent: Hy3$0.13/1M
- 3Tencent: Hunyuan A13B Instruct$0.14/1M
Highlights
Quality
Overall capability · Higher is better
Speed
Output tokens per second · Higher is better
Input price
USD per 1M tokens · Lower is better
Quality benchmarks
Compare Tencent model capability.
Performance and price
Compare practical tradeoffs.
All Tencent models.
3 models in the current catalogue.
| Model | Links | |||||
|---|---|---|---|---|---|---|
| Tencent: Hy3 preview | 41.2 | 27 t/s | 262.1K | $0.06 | $0.21 | |
| Tencent: Hunyuan A13B Instruct | 0 | 2 t/s | 131.1K | $0.14 | $0.57 | |
| Tencent: Hy3 | 0 | 51 t/s | 262.1K | $0.13 | $0.53 |
Tencent technical benchmarks.
Compare named evaluation records available for Tencent models. Missing results remain blank rather than being estimated.
| Benchmark | Area | Hy3 preview | Hunyuan A13B Instruct | Hy3 | Source |
|---|---|---|---|---|---|
| Intelligence Index | overallQuality | 41.2 | — | — | OpenRouter / artificial-analysis |
| Coding Index | coding | 58.8 | — | — | OpenRouter / artificial-analysis |
| Agentic Index | agentic | 30.7 | — | — | OpenRouter / artificial-analysis |
| GPQA | GPQA Diamond | 89.7 | — | — | OpenRouter model benchmarks |
| Humanity's Last Exam | HLE | 31.6 | — | — | OpenRouter model benchmarks |
| AA-LCR | AA-LCR | 66.7 | — | — | OpenRouter model benchmarks |
| GDPval-AA | GDPval-AA | 35.8 | — | — | OpenRouter model benchmarks |
| CritPt | CritPt | 4.9 | — | — | OpenRouter model benchmarks |
| SciCode | SciCode | 47.6 | — | — | OpenRouter model benchmarks |
| AA-Omniscience Accuracy | AA-Omniscience Accuracy | 31.5 | — | — | OpenRouter model benchmarks |
| AA-Omniscience Non-Hallucination Rate | AA-Omniscience Non-Hallucination Rate | 27 | — | — | OpenRouter model benchmarks |