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-21 Sep 2026
Status
Older
Current
Open weightsYes, Modified MIT LicenseNo
Hosts
APIs you can call it on
18
Price · USD per million tokens, at the lab's own API where it sells the model
Priced atFireworks AIxAI
Input$2.00$2.00
Output$8.00$6.00
Cached input$0.30$0.50
Batch
input / output
-$1.60 / $4.80
Long prompts
input / output
-$4.00 / $12.00
above 200K
Cheapest anywhere
input / output
$2.00 / $8.00
Fireworks AI
$2.00 / $6.00
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
$1,047$915
RAG over documents
2,000 a day, 8K in, 600 out
$840$816
Coding agent
1,000 a day, 40K in, 2K out
$1,452$1,500
Overnight batch
20,000 a day, 4K in, 400 out
$6,720$4,992
Limits
Context window262K tokens500K tokens
Max output33K tokens500K tokens
Capabilities
Takes inText, ImagesText, Images
ProducesTextText
Effort settings
as listed by the source
--
Streaming
Function calling
JSON mode
Structured outputs
Reasoning
Tool calling
Prompt caching
Batch API
Built-in web search