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
Granite Speech 3.3 8B
IBM · Transcription
Overview
Released-
Status
Current
Open weightsYes, Apache-2.0
Hosts
APIs you can call it on
1
Price · USD per million tokens, at the lab's own API where it sells the model
Priced atReplicate
Input$0.00
Output$4.90
Cached input-
Batch
input / output
-
Cheapest anywhere
input / output
$0.00 / $4.90
Replicate
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
$368
RAG over documents
2,000 a day, 8K in, 600 out
$176
Coding agent
1,000 a day, 40K in, 2K out
$294
Overnight batch
20,000 a day, 4K in, 400 out
$1,176
Limits
Context window-
Max output-
Capabilities
Takes inText
ProducesText
Effort settings
as listed by the source
-