Skip to content
Dashboard

Relace models

Relace is a hosted inference provider for open-weight AI models, delivering cost-efficient access optimized for coding, reasoning, and agentic workloads.
Model
Context
Input
Output
Latency
Throughput
Cache
Web Search
Capabilities
ZDR
No Training
Regional Inference
Free Tier
Released
1M
$0.10/M
$0.50/M
4.8 s30 tps
Read$0.01/M
—
+1
09/08/2026
1M
$0.07/M
$0.28/M
0.8 s58 tps
Read$0.02/M
—
+1
08/26/2026
1M
$0.03/M
$0.32/M
1.3 s32 tps
Read$0.02/M
—
+1
07/31/2026