GPT-5.5 vs Hy3 preview
Price, context and benchmark scores side by side — every number from the live model catalogue.
| GPT-5.5openai | Hy3 previewtencent | |
|---|---|---|
| Input / 1M tokens | $5.00 | $0.063 |
| Output / 1M tokens | $30.00 | $0.210 |
| 1M in + 300K out | $14.00 | $0.13 |
| Context window | 1.1M | 262K |
| Intelligence index | 54.8 | 41.2 |
| Coding index | 74.9 | 58.8 |
| Agentic index | 44.9 | 30.7 |
| Cheapest provider | Azure | GMICloud |
| Providers serving it | 2 | 1 |
| Weights | Closed | Open |
| Image input | Yes | No |
| Tool calling | Yes | Yes |
| Reasoning mode | Yes | Yes |
| Released | Apr 24, 2026 | Apr 22, 2026 |
Which one to pick
Pick GPT-5.5 if…
- You want the higher general-intelligence score (54.8).
- Code quality is the priority — it leads on the coding index (74.9).
- You are running agent loops — higher agentic index (44.9).
- You need the bigger context window (1.1M).
- You want provider redundancy: 2 providers serve it.
Pick Hy3 preview if…
- Token cost matters: $0.210 per 1M output vs $30.00.
- You want the option to self-host — weights are public.
FAQ
Is GPT-5.5 better than Hy3 preview?
GPT-5.5 scores higher on general intelligence (54.8 vs 41.2), while Hy3 preview is 143× cheaper per output token ($0.210 vs $30.00). For code, GPT-5.5 leads (74.9 coding index). GPT-5.5 takes the larger context at 1.1M.
Which is cheaper, GPT-5.5 or Hy3 preview?
Hy3 preview, at $0.210 per 1M output tokens versus $30.00. On a 1M input + 300K output workload that is $0.13 against $14.00.
Can I self-host GPT-5.5 or Hy3 preview?
GPT-5.5: closed weights, API only. Hy3 preview: open weights (tencent/Hy3-preview).