Skip to content
Dashboard

Inference.net models

Inference.net is an AI inference platform for deploying and scaling open-source, custom, and fine-tuned language models. It combines managed model hosting with production monitoring, agent tracing, evaluation, and training, helping teams optimize quality, latency, and cost for AI agents, data extraction, and classification workflows.
Model
Context
Input
Output
Latency
Throughput
Cache
Search
Capabilities
ZDR
No Training
Regional Inference
Free AI Gateway Credit
Released
1M
$0.14/M
$0.28/M
——
Read$0.0028/M
—
+2
09/21/2026
1M
$0.30/M
$1/M
0.8 s218 tps
Read$0.007/M
—
+2
09/08/2026
1M
$0.10/M
$0.45/M
2.1 s94 tps
Read$0.03/M
—
+2
08/26/2026
1M
$0.90/M
$3/M
1.0 s212 tps
Read$0.15/M
—
+1
08/18/2026
1M
$3/M
$15/M
1.2 s130 tps
Read$0.30/M
—
+2
07/16/2026
128K
$0.05/M
$0.23/M
1.0 s24 tps
Read$0.05/M
—
04/16/2026
128K
$0.03/M
$0.15/M
1.0 s—
Read$0.03/M
—
04/16/2026