LiquidAI LFM2.5-24B-A2B vs Tencent Hy3 preview
Side by side from live data: prices, limits, who serves each model and how reliably, and how they do on our chess arena.
In short
- LFM2.5-24B-A2B is 1.3× cheaper at a typical mix of 3 input tokens per output token.
- Hy3 preview accepts longer prompts: 262K vs 128K tokens.
- LFM2.5-24B-A2B's best provider was more reliable over 24 h: 100% vs 97.8%.
- LFM2.5-24B-A2B is the newer model (released 2026-10-13).
| LFM2.5-24B-A2B | Hy3 preview | |
|---|---|---|
| Vendor | LiquidAI | Tencent |
| Released | 2026-10-13 | 2026-04-22 |
| Input / 1M | $0.05 | $0.18 |
| Output / 1M | $0.70 | $0.60 |
| Cached input / 1M | $0.025 | $0.06 |
| Blended / 1M (3:1) | $0.212 | $0.285 |
| Context | 128K | 262K |
| Max output | 8K | 236K |
| Accepts | text | text |
| Reasoning | Yes | Yes |
| Providers on OpenRouter | 1 | 1 |
| Cheapest provider, output | $0.70 | $0.60 |
| Best provider uptime, 24 h | 100% | 97.8% |
| Chess arena rating | not playing | not playing |
| Vendor API status | — | — |
Cost calculator
LFM2.5-24B-A2B
$13.50 per month$0.45 per day
Hy3 preview
$19.80 per month$0.66 per day
LFM2.5-24B-A2B saves $6.30 a month on this workload.
At list prices per 1M tokens; prompt caching and batch discounts are not included.
Compare LFM2.5-24B-A2B with