Skip to content
NewTry RouterPlus with free creditsTry playground →
RouterPlus
Book a demo

Morph API

Open in consoleI need inference
  • 8Models
  • 4Labs
  • 1.05MMax context
  • $0.12/ $0.70From, per 1M in / out

Pricing

ModelLabContextInput / 1MCached / 1MOutput / 1MProviders
DeepSeek V4.1 Flashdeepseek/deepseek-v4.1-flashDeepSeek1.05M$0.12$0.008$0.809
GLM 5.3 Flashz-ai/glm-5.3-flashZ.ai1.05M$0.20$0.04$0.7012
GLM 5.3z-ai/glm-5.3Z.ai1.05M$1.19$0.20$6.0012
Kimi K3moonshotai/kimi-k3Moonshot AI1.05M$2.50$0.29$14.0010
Kimi K3 Fastmorph/kimi-k3-fastMorph1.05M$6.00$0.60$22.501
GLM 5.2z-ai/glm-5.2Z.ai1.05M$1.40$0.26$4.408
Morph v3 Fast Applymorph/morph-v3-fastMorph262K$0.80–$1.201
Morph v3 Large Applymorph/morph-v3-largeMorph262K$0.90–$1.901

Morph's price, USD, as of November 5, 2026. Live prices

Quickstart

OpenAI-compatible
curl https://api.routerplus.com/v1/chat/completions \
  -H "Authorization: Bearer $ROUTERPLUS_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "z-ai/glm-5.3",
    "messages": [{ "role": "user", "content": "Hello" }],
    "max_tokens": 1024,
    "provider": { "only": ["morphllm"] }
  }'

FAQ

How do I send a request to Morph?

Routing tries providers cheapest first, by input plus output price, so Morph serves a model where it is the cheapest and otherwise as a fallback. To send a request to Morph first, name it in provider.only or provider.order as "morphllm".

How much does Morph cost on RouterPlus?

You pay Morph's own price for the model. Morph's lowest input price is $0.12 per million tokens (DeepSeek V4.1 Flash) and its lowest output price is $0.70 per million tokens (GLM 5.3 Flash). Prices as of November 5, 2026.

Which models does Morph serve?

DeepSeek V4.1 Flash, GLM 5.3 Flash, GLM 5.3, Kimi K3, Kimi K3 Fast, GLM 5.2, Morph v3 Fast Apply and Morph v3 Large Apply.

Does Morph keep my prompts?

Morph declares that it retains prompts. It offers zero data retention only for self-hosted deployments, not its hosted API.

Where does Morph run inference?

In the US, and DeepSeek V4.1 Flash in Japan.

How do I keep Morph's cache warm?

Send a session_id (or a prompt_cache_key). The gateway passes it to Morph as the x-session-id header, which Morph uses to pick the region and worker that hold your cache.

Route to Morph with one key

Open in consoleI need inference