Model facet · Providers

GLM-5.3-Flash inference providers

GLM-5.3-Flash is served through Z.ai first-party, Novita, and Together AI on OneInfer, with the OpenRouter listing as the verified reference. Z.ai and Novita both list at $0.075 input and $0.250 output per 1M tokens; Together AI's `zai-org/GLM-5.3-Flash` route lists at 2× that price ($0.15 input / $0.50 output).

Tracked providers

ProviderInput /1M tokensOutput /1M tokensProvenance
Z.ai$0.075$0.250vendor reported
Novita$0.075$0.250vendor reported
Together AI$0.150$0.500vendor reported

What is GLM-5.3-Flash?

GLM-5.3-Flash is Z.ai's natively multimodal mixture-of-experts model in the GLM-5 family — 320B total parameters with 18B active per token, BF16 weights, 1,048,576-token context window, image + text input, text output. Released 26 Aug 2026, priced at $0.075 per 1M input tokens and $0.250 per 1M output tokens on the OpenRouter listing.

Providers

Each row is labeled by provenance: independently verified listings versus a provider's own vendor-reported number. Routing, quantization, and uptime can all change — record the provider and date alongside any benchmark or latency claim you rely on.

Ready to test the workflow?

Create account & add credits

When GLM-5.3-Flash is not the right choice

GLM-5.3-Flash is not the cheapest flash-tier option on input or output — DeepSeek V4 Flash 0731 is 2.5× cheaper on input and 3.3× cheaper on output. The defensive position for GLM-5.3-Flash is native multimodal (image + text input) at flash-tier pricing. If you need only text, DeepSeek V4 Flash 0731 is the cheaper option; if you need stronger reasoning on coding agents, the Qwen3.8-Max tier is the upgrade path.

Frequently asked questions

Where is the live GLM-5.3-Flash model page?

The canonical model page with current OneInfer pricing, capabilities, and availability is /models/zai-org/GLM-5.3-Flash. This page is a focused facet of that entity, not a replacement for it.

How should I treat benchmark or price claims?

Check each claim's provenance label and observed date. Vendor-reported and independently verified numbers are shown as separate evidence classes on this hub.

Put GLM-5.3-Flash to work

Fund a controlled evaluation, start with a prepared prompt, and measure quality and cost on your own workload.