llmvision/glimpse-v1

llmvision/ glimpse-v1

27 Downloads Updated yesterday

Private visual Intelligence for your home.

vision

ollama run llmvision/glimpse-v1

curl http://localhost:11434/api/chat \
  -d '{
    "model": "llmvision/glimpse-v1",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='llmvision/glimpse-v1',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'llmvision/glimpse-v1',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Models

View all →

Name

1 model

Size

Context

Input

glimpse-v1:latest

5.0GB · 128K context window · Text, Image · yesterday

glimpse-v1:latest

5.0GB

128K

Text, Image

Readme

Glimpse-v1 is a lightweight vision-language model (VLM) built to summarize home security camera events. It natively supports structured JSON, enabling seamless integration into automations.

Get Started

Install Ollama
Set up LLM Vision for Home Assistant
Pull this model using the command at the top of this page.
Set up the Ollama provider in LLM Vision
Choose llmvision/glimpse-v1 as the default model

Benchmark

The following benchmark compares semantic similarity (CLIP) of a validation response to the model’s response.

Mean Score Overall

Higher is better

Mean Score by Category Higher is better

Latency Lower is better

Latency vs Score Lower is better

**Glimpse-v1** is a lightweight vision-language model (VLM) built to summarize home security camera events. It natively supports structured JSON, enabling seamless integration into automations.

### Get Started
1. Install Ollama
2. Set up [LLM Vision](https://llmvision.org/) for Home Assistant
3. Pull this model using the command at the top of this page.
4. [Set up the Ollama provider](https://llmvision.gitbook.io/getting-started/setup/providers/self-hosted-providers#ollama) in LLM Vision
5. Choose `llmvision/glimpse-v1` as the default model

### Benchmark
The following benchmark compares semantic similarity (CLIP) of a validation response to the model's response.

**Mean Score Overall**

![mean_score_overview.png](/assets/llmvision/glimpse-v1/d2bed95f-7e0d-42a5-adc6-ff0efc0bb512)_Higher is better_

**Mean Score by Category**
![mean_score_categories.png](/assets/llmvision/glimpse-v1/3077249c-1075-43d8-806d-b161f089766c)_Higher is better_

**Latency**
![latency.png](/assets/llmvision/glimpse-v1/e39b22db-64b3-479b-8278-950d51750345)_Lower is better_

**Latency vs Score**
![latency_vs_score.png](/assets/llmvision/glimpse-v1/e21a3753-4ba4-4f58-a385-573c5a04259b)_Lower is better_

Paste, drop or click to upload images (.png, .jpeg, .jpg, .svg, .gif)