ZMIME
Vispark · Released May 15, 2024

Vision Small

Fast, low-cost multimodal model for understanding text, images, audio, video, and PDFs, with tool calling and a 1M-token context window.

  • Proprietary
  • Reasoning
  • Tool calling
  • Vision
  • Audio input
  • Video input
Capability
—No ECI score yet
Input
$1.05per 1M tokens
Output
$3.16per 1M tokens
Context
1M66K max output
Pricing

What it costs

Input
$1.05per million tokens
Output
$3.16per million tokens
Blended (3:1)
$1.58pricier than 68% of priced models

Typical monthly bills

Monthly usageEstimated cost
Side project2M input + 0.5M output tokens$3.68
Team assistant25M input + 5M output tokens$42.05
Production app250M input + 50M output tokens$420.50

Official Vispark API price, as listed on models.dev.

Capability

Independent benchmarks

Epoch AI has not published a Capabilities Index score for Vision Small yet.

Specifications

The details

Lab
Vispark
Released
May 15, 2024
Context window
1,000,000 tokens
Max output
65,536 tokens
Inputs
Text, Images, PDFs, Audio, Video
Output
Text
Reasoning
Always on
Tool calling
Yes
Structured output
Yes
Weights
Proprietary
API model ID
vispark/vision-smallon Vispark
Availability
1 API providerlisted on models.dev

Vispark model documentation

Same lab

More from Vispark

Questions

About Vision Small

How much does Vision Small cost?

Vision Small costs $1.05 input / $3.16 output per million tokens (official Vispark API price). At a 3:1 input-to-output mix that is $1.58 per million tokens, more expensive than 68% of the 469 priced models we track.

What is the context window of Vision Small?

Vision Small accepts up to 1,000,000 tokens per request and can write up to 65,536 tokens in one response.

How good is Vision Small?

Epoch AI has not published a Capabilities Index score for Vision Small yet.

Is Vision Small open source?

No. Vision Small is proprietary; you use it through Vispark’s API or partner platforms.

What inputs does Vision Small support?

Vision Small accepts text, images, PDFs, audio and video and replies in text. It is a reasoning model, supports tool calling and can return structured JSON output.

When was Vision Small released?

Vispark released Vision Small on May 15, 2024.

What are the best alternatives to Vision Small?

The closest current models from other labs on capability, price and release date are Qwen Max (Alibaba (Qwen), $1.60 / $6.40), Hermes 3 405B Instruct (Nous Research, $1.00 / $1.00), Nova Pro (Amazon, $0.80 / $3.20) and Codestral-22B-v0.1 (Mistral AI).