Model overview
gpt-oss-120b是OpenAI的开放权重、117b参数的专家混合（MoE）语言模型，专为高推理、代理和通用生产用例而设计。它每次向前传递激活5.1亿个参数，并经过优化，可以在单个H100 GPU上运行，具有本机MXFP4量化。该模型支持可配置的推理深度、完整的思维链访问和本地工具使用，包括函数调用、浏览和结构化输出生成。
The summary below is compiled from sources by an AI model, not our own testing. Verify selection decisions against original sources and your workload.
Key features
- 采用专家混合（MoE）架构，117B总参数，每次前向传递仅激活5.1亿参数，兼顾规模与效率
- 支持本机MXFP4量化，可在单个H100 GPU上运行，降低硬件部署门槛
- 提供可配置推理深度、完整思维链访问及本地工具调用能力（含函数调用、浏览、结构化输出生成）
Limitations & considerations
Sources are insufficient to establish specific limitations. Verify task performance, latency, data policies, licensing and version lifecycle.
Source prices & cost analysis
Prices vary by channel, region and configuration. Channel quotes are listed independently and are not treated as a single official provider price.
| Source / channel | Input | Output | Billing unit | Collected at |
|---|---|---|---|---|
| OpenRouter · openai/gpt-oss-120b ↗ | USD 0.019 | USD 0.069 | Per million tokens | 2026/11/2 06:25:30 |
| Apilio · AZ-优质 ↗ | Not verified | Not verified | 渠道倍率待核验 | 2026/11/2 06:25:15 |
| Apilio · default ↗ | Not verified | Not verified | 渠道倍率待核验 | 2026/11/2 06:25:15 |
| Apilio · openai官-优质 ↗ | Not verified | Not verified | 渠道倍率待核验 | 2026/11/2 06:25:15 |
| Apilio · origin ↗ | Not verified | Not verified | 渠道倍率待核验 | 2026/11/2 06:25:15 |
| Apilio · ssvip ↗ | Not verified | Not verified | 渠道倍率待核验 | 2026/11/2 06:25:15 |
| Apilio · svip ↗ | Not verified | Not verified | 渠道倍率待核验 | 2026/11/2 06:25:15 |
Some channels offer cache discounts. Check the source for cache read, write and additional fees.
Estimate monthly base token costs
Estimates exclude cache discounts, tiered pricing, multimodal surcharges, web search, taxes and additional reasoning output.
Public benchmarks & run configuration
暂无已采集的可核验公开评分。模型未上榜不代表能力较低。
View full indexed rankings →Frequently asked questions
gpt-oss-120b是否为OpenAI官方发布的模型？
资料中称其为'OpenAI的开放权重模型'，但未提供OpenAI官方发布声明或仓库链接；所有引用来源均为第三方平台（Apilio、OpenRouter、Hugging Face等），真实性未经OpenAI验证。
该模型是否支持长上下文输入？
资料明确标注context=131072，表明最大上下文长度为131,072 token。
它能否在消费级GPU（如RTX 4090）上运行？
资料仅声明'可在单个H100 GPU上运行'，未提及对RTX 4090或其他消费级GPU的支持情况。
Do open weights permit free commercial use?
Not necessarily. Check the version license, commercial conditions and usage policies. Self-hosting incurs compute and maintenance costs.
Are these real-time billing prices?
These are timestamped source snapshots, not live billing prices. Check the channel for current versions, regions, discounts and billing methods.
Sources & traceability
许可证：apache-2.0。模型介绍与第三方评测均以原来源的具体版本和方法为准。