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-31 Mar 2026
Status
Current
in 131 days Leaves Amazon Bedrock on 30 Mar 2027
Current
Open weightsYes, DeepSeek License (MIT code)No
Hosts
APIs you can call it on
14
Price · USD per million tokens, at the lab's own API where it sells the model
Priced atAmazon BedrockxAI
Input$1.35$1.25
Output$5.40$2.50
Cached input-$0.20
Batch
input / output
-$1.00 / $2.00
Long prompts
input / output
-$2.50 / $5.00
above 200K
Cheapest anywhere
input / output
$1.35 / $5.40
Amazon Bedrock
$1.25 / $2.50
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
$810$468
RAG over documents
2,000 a day, 8K in, 600 out
$842$438
Coding agent
1,000 a day, 40K in, 2K out
$1,944$768
Overnight batch
20,000 a day, 4K in, 400 out
$4,536$2,880
Limits
Context window128K tokens1M tokens
Max output8K tokens1M tokens
Capabilities
Takes inTextText, Images
ProducesTextText
Effort settings
as listed by the source
--
Streaming
JSON mode
Structured outputs
Reasoning
Prompt caching
Batch API
Built-in web search