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
Released16 Oct 202618 Jul 2024
Status
Current
Current
in 56 days Leaves OpenAI on 14 Jan 2027
Open weightsYes, Apache-2.0No
Hosts
APIs you can call it on
25
Price · USD per million tokens, at the lab's own API where it sells the model
Priced atW&B InferenceOpenAI
Input$0.30$0.15
Output$1.00$0.60
Cached input$0.04$0.075
Batch
input / output
-$0.075 / $0.30
Cheapest anywhere
input / output
$0.30 / $1.00
W&B Inference
$0.15 / $0.60
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
$142$83.25
RAG over documents
2,000 a day, 8K in, 600 out
$118$75.60
Coding agent
1,000 a day, 40K in, 2K out
$202$153
Overnight batch
20,000 a day, 4K in, 400 out
$960$252
Limits
Context window1M tokens128K tokens
Max output1M tokens16K tokens
Capabilities
Takes inTextText, Images, PDFs
ProducesTextText
Effort settings
as listed by the source
--
Streaming
Function calling
JSON mode
Structured outputs
Reasoning
Tool calling
Prompt caching
Batch API
Parallel tool calls