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-20 May 2026
Status
Current
Current
Open weightsYes, DeepSeek License (MIT code)No
Hosts
APIs you can call it on
12
Price · USD per million tokens, at the lab's own API where it sells the model
Priced atEmpirioLabs AIxAI
Input$1.65$1.00
Output$3.30$2.00
Cached input$0.138$0.20
Batch
input / output
--
Long prompts
input / output
-$2.00 / $4.00
above 200K
Cheapest anywhere
input / output
$1.65 / $3.30
EmpirioLabs AI
$1.00 / $2.00
xAI
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
$606$378
RAG over documents
2,000 a day, 8K in, 600 out
$548$360
Coding agent
1,000 a day, 40K in, 2K out
$908$648
Overnight batch
20,000 a day, 4K in, 400 out
$4,752$2,880
Limits
Context window1M tokens256K tokens
Max output393K tokens256K tokens
Capabilities
Takes inTextText, Images
ProducesTextText
Effort settings
as listed by the source
--
Streaming
Function calling
JSON mode
Structured outputs
Reasoning
Tool calling
Prompt caching
Built-in web search