impulse

Compare models

Up to 4 models side by side: list prices, what a month of real traffic costs, limits and capabilities. The best value in each row is marked with a check.

Clear
Model
Overview
Released-24 Feb 2026
Status
Current
Current
in 133 days Leaves OpenAI on 1 Apr 2027
Open weightsNoNo
Hosts
APIs you can call it on
15
Price · USD per million tokens, at the lab's own API where it sells the model
Priced atVisparkOpenAI
Input$4.21$1.75
Output$12.63$14.00
Cached input$0.00$0.175
Batch
input / output
--
Cheapest anywhere
input / output
$4.21 / $12.63
Vispark
$1.75 / $14.00
OpenAI
Monthly bill · at the price above, 30 days, with each workload's cache and batch use
Chat assistant
5,000 a day, 2K in, 500 out
$1,831$1,433
RAG over documents
2,000 a day, 8K in, 600 out
$1,465$966
Coding agent
1,000 a day, 40K in, 2K out
$2,273$1,617
Overnight batch
20,000 a day, 4K in, 400 out
$13,135$7,560
Limits
Context window1M tokens272K tokens
Max output66K tokens128K tokens
Capabilities
Takes inAudio, Images, Text, Video, PDFsImages, Text, PDFs
ProducesTextText
Effort settings
as listed by the source
-none, xhigh
Streaming
Function calling
JSON mode
Structured outputs
Reasoning
Tool calling
Prompt caching
Built-in web search
Parallel tool calls
Tool search