Skip to content
Dashboard

Boundless models

Boundless is an AI inference platform and engineering partner helping startups run open-weight models in production. Its engineers select models for each workload, tune serving infrastructure to real traffic, and manage ongoing operations to improve throughput and reduce costs. Teams get inference tailored to their product, with hands-on support as usage grows and better models emerge.
Model
Context
Input
Output
Latency
Throughput
Cache
Web Search
Capabilities
ZDR
No Training
Regional Inference
Free Tier
Released
1M
$0.20/M
$1/M
0.5 s58 tps
Read$0.01/M
+1
09/08/2026
1M
$0.12/M
$0.40/M
3.8 s162 tps
Read$0.02/M
+1
08/26/2026