SIDE BY SIDE
gpt-oss-120b vs Hy3 preview
Each is the newest model of its family in Blossom Valley, and the two are neighbours there by blended price.
| Field | gpt-oss-120b | Hy3 preview |
|---|---|---|
| Provider | OpenAI | Tencent |
| Family | GPT | Tencent |
| Province | Blossom Valley | Blossom Valley |
| Context window | 131K tokens | 262K tokens |
| Maximum output | 66K tokens | 236K tokens |
| Input / 1M tokens | $0.150 | $0.180 |
| Output / 1M tokens | $0.600 | $0.600 |
| Accepts | text | text |
| Produces | text | text |
| Reasoning | Yes | Yes |
| Tool use | Yes | Yes |
| Added to OpenRouter | 2025-08-05 | 2026-04-22 |
| Model ID | openai/gpt-oss-120b | tencent/hy3-preview |
Questions and answers
Which is cheaper, gpt-oss-120b or Hy3 preview?
gpt-oss-120b costs less for both input ($0.150 vs $0.180) and output ($0.600 vs $0.600) per million tokens. On input, gpt-oss-120b is 17% cheaper.
Which has the larger context window?
Hy3 preview: 262K tokens against 131K for gpt-oss-120b (2× as long).
What can one do that the other cannot?
Nothing among reasoning, tool use and image input: both support reasoning and tool use.
Answers are computed from the 2026-10-24 catalog snapshot.