Model overview
Glimpse is a multimodal visual reasoning model from Elorian, built for visual understanding and complex visual reasoning in the physical world. It reasons directly over images and video rather than...
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
- 支持多模态输入（文本、图像、视频）进行视觉理解与物理世界中的复杂视觉推理
- 直接在原始图像和视频上进行推理，不依赖中间表征或预提取特征
- 支持结构化输出，适用于需明确逻辑或工具调用的推理任务
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 · elorian/glimpse ↗ | USD 0.6 | USD 3 | Per million tokens | 2026/11/2 06:25:21 |
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
Glimpse 是否支持视频输入？
是的，Glimpse 明确支持视频作为输入模态之一。
Glimpse 的上下文长度是多少？
其上下文长度为 262144（tokens 或等效处理单元，具体单位未在资料中说明）。
Glimpse 是否能生成非文本输出（如图像或音频）？
否，资料明确指出其输出仅为文本。
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：来源未披露，须进一步核验。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。