Speed-optimised inference for coding agents — US-hosted.
Morph runs an inference platform tuned for coding agents, serving its own Fast Apply edit models (morph-v3-fast, morph-v3-large) alongside speed-optimised open-weight models from DeepSeek, MiniMax, and Qwen. US-hosted. The right route when low-latency code edits and high-throughput agent loops matter and US hosting is acceptable.
Every model we route through Morph. Compare residency, zero data retention, training posture and price at a glance, with full data-handling detail per route below.
| Model | Region | Zero data retention | Training | Context | Input | Output |
|---|---|---|---|---|---|---|
| US | No | 1M | $1.00 | $3.41 | ||
| US | No | 1M | $2.50 | $14.00 | ||
| US | No | 1M | $6.00 | $22.50 | ||
| US | No | 1M | $0.10 | $0.35 | ||
| US | No | 1M | $0.30 | $1.20 | ||
| US | No | 1M | $0.12 | $0.35 | ||
| US | No | 1M | $0.12 | $0.35 | ||
| US | No | 1M | $0.12 | $0.35 | ||
| US | No | 1M | $1.10 | $4.10 | ||
| US | No | 256K | $0.60 | $2.40 | ||
| US | No | 197K | $0.28 | $1.20 | ||
| US | No | 262K | $0.50 | $3.50 | ||
| US | No | 131K | $0.29 | $2.40 |
Morph hosts on 1 route. Each route has its own privacy posture, residency, and GDPR terms. Postures are maintained by Opper with a last-verification timestamp.
Zero data retention with abuse monitoring: prompts and outputs are kept for up to 30 days for abuse monitoring only, and are not used for training. No training on customer data. US; SCCs; DPA available.
