Vercel News

Elorian AI's Glimpse now available on AI Gateway

8.5内容质量

TL;DR · AI 摘要

Elorian AI的Glimpse模型现已在AI Gateway上提供,支持图像和视频的视觉推理任务。

核心要点

  • Glimpse模型支持图像和视频分析,适用于科学图表分析、视频字幕生成等场景。
  • 使用AI SDK时,需指定模型名为elorian/glimpse,并可发送视频文件进行处理。
  • AI Gateway提供统一API,支持自定义报告、零数据保留和预算管理等功能。

结构提纲

按章节快速跳转。

  1. 宣布Elorian AI的Glimpse模型在AI Gateway上线。

  2. Glimpse模型能处理图像和视频,分析运动轨迹与空间关系。

  3. 展示科学图表分析、视频字幕生成等具体应用场景。

  4. 提供AI SDK代码示例,说明如何调用模型处理文件。

  5. 介绍通过Vercel CLI配置AI Gateway的步骤。

  6. AI Gateway提供统一API、成本追踪和性能优化功能。

思维导图

用一张图看清主题之间的关系。

查看大纲文本(无障碍 / 无 JS 友好)
  • Elorian Glimpse模型
    • 核心功能
      • 视觉推理(图像/视频)
      • 运动轨迹分析
      • 空间关系解读
    • 技术实现
      • AI SDK调用示例
      • 视频文件处理
    • AI Gateway特性
      • 统一API管理
      • 零数据保留
      • 预算控制

金句 / Highlights

值得收藏与分享的关键句。

#AI模型#视觉推理#AI Gateway#Vercel
打开原文

Elorian AI's Glimpse now available on AI Gateway - Vercel

Glimpse from Elorian is now available on AI Gateway. The visual reasoning model accepts images and video to follow objects through motion, interpret spatial relationships, and connect an action to its physical consequences.

Glimpse can return to visual evidence as it works: zooming into a detail, revisiting a short interval of video, or tracking the same feature across frames. This allows the model to support use cases like scientific chart analysis, video captioning, visual counting, and robotics data annotation.

Some examples include checking a train's motion to determine whether it is slowing down, keeping objects distinct through camera movement, and labeling completed actions in robot footage.

Use elorian/glimpse as the model name:

AI SDK

Chat Completions

code
1
import
{
streamText
}
from
'ai'
;
2
3
const
result
=
streamText
(
{
4
model
:
'elorian/glimpse'
,
5
messages
:
[
{
6
role
:
'user'
,
7
content
:
[
8
{
type
:
'text'
,
text
:
'Describe this image in one sentence.'
}
,
9
{
10
type
:
'file'
,
11
mediaType
:
'image/jpeg'
,
12
data
:
new
URL
(
'https://elorian.ai/assets/blog/glimpse/open-pliers.jpg'
)
,
13
}
,
14
]
,
15
}
]
,
16
maxOutputTokens
:
4096
,
17
reasoning
:
'low'
18
}
)
;
19
20
for
await
(
const
text
of
result
.
textStream
)
{
21
process
.
stdout
.
write
(
text
)
;
22
}

Copy for agent

View model details

↗

With AI SDK 7, you can send a video as a file part. Replace the placeholder URL with a public or signed HTTPS URL for your video:

code
1
import
{
streamText
}
from
'ai'
;
2
3
const
result
=
streamText
(
{
4
model
:
'elorian/glimpse'
,
5
maxOutputTokens
:
4096
,
6
messages
:
[
{
7
role
:
'user'
,
8
content
:
[
9
{
type
:
'text'
,
text
:
'Describe the visible scene in at most five words. Do not use code or search.'
}
,
10
{
11
type
:
'file'
,
12
data
:
new
URL
(
'https://example.com/your-video.mp4'
)
,
13
mediaType
:
'video/mp4'
,
14
}
,
15
]
,
16
}
]
,
17
reasoning
:
'low'
,
18
providerOptions
:
{
elorian
:
{
video_fps
:
1
}
}
19
}
)
;
20
21
for
await
(
const
text
of
result
.
textStream
)
{
22
process
.
stdout
.
write
(
text
)
;
23
}

To configure AI Gateway in Claude Code, Codex, Cursor, and more, install the latest Vercel CLI and run setup:

code
npx vercel ai-gateway setup

Try Glimpse in the model playground .

AI Gateway provides a unified API for calling models, tracking usage and cost, and configuring retries, failover, and performance optimizations for higher-than-provider uptime. It includes built-in custom reporting , Zero Data Retention support , budgets for API keys , routing rules , and more.

AI Gateway reflects provider pricing with no markup and does not charge a platform fee on inference, including on Bring Your Own Key (BYOK) requests.