GLM 5.3 Flash API
| Provider | Input / 1M | Cached / 1M | Output / 1M |
|---|---|---|---|
| Morph | $0.20 | $0.04 | $0.70 |
| DeepInfra | $0.15 | $0.03 | $0.50 |
| Nebius | $0.15 | $0.03 | $0.50 |
| Novita AI | $0.15 | $0.03 | $0.50 |
| Inference.net | $0.10 | $0.03 | $0.45 |
| RunInfra | $0.11 | $0.03 | $0.45 |
USD, as of October 27, 2026. Live prices
| Workload | Morph | Inference.net |
|---|---|---|
| 1,000 chat replies2K in, 500 out each | $0.75 | $0.43 |
| 1,000 agent tasks50K in, 5K out each | $13.50 | $7.25 |
| Monthly batch job500M in, 50M out | $135.00 | $72.50 |
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-flash",
"messages": [{ "role": "user", "content": "Hello" }]
}'On RouterPlus, routing for GLM 5.3 Flash starts at Morph: $0.20 per million input tokens and $0.70 per million output tokens. The lowest listed prices are at Inference.net: $0.10 input and $0.45 output. Name a provider with provider.only and you pay its price. Prices as of October 27, 2026.
6 providers serve GLM 5.3 Flash on RouterPlus: Morph, DeepInfra, Nebius, Novita AI, Inference.net and RunInfra. By default the router tries them in routing order and fails over before the first byte, never mid-answer.
Use "z-ai/glm-5.3-flash" as the model with the OpenAI-compatible endpoint at https://api.routerplus.com/v1. The same key also accepts the Anthropic API format.
1.05M tokens of context and up to 524K output tokens, per the RouterPlus catalog. A provider can list a lower limit; the console shows each one.
Yes. Z.ai publishes GLM 5.3 Flash as an open-weight model, which is why several independent providers can host it.
5 of 6 providers list no prompt retention (DeepInfra, Nebius, Novita AI, Inference.net and RunInfra). Morph may retain prompts. Pin the providers you accept with a routing policy.
Yes. Tell us the model, the throughput you need and the region with "I need inference", and RouterPlus lines up reserved capacity for it.
More GLM