GLM Free API – $0 GLM-5.3 via TokenRouter, Huawei Cloud

TokenRouter lists the separate z-ai/glm-5.3-free model at $0.0000 per 1M input tokens and $0.0000 per 1M output tokens through its OpenAI-compatible chat completions API. This is model pricing rather than a coupon or one-time credit. TokenRouter does not publish a fixed expiry date or a specific free-model usage quota on the offer page, so those limits could not verify.

Z.ai’s current official pricing lists GLM-4.7-Flash, GLM-4.5-Flash, and GLM-4.6V-Flash as free for input, cached input, cached-input storage, and output. GLM-4.7-Flash and GLM-4.5-Flash are text models, while GLM-4.6V-Flash accepts video, image, text, and file inputs. This is actual $0 model pricing rather than a one-time credit. However, “free” should not be read as unlimited: Z.ai documents API rate-limit and daily-limit errors, while a fixed universal quota for these specific free models could not be verified.

Huawei Cloud CodeArts gives each user 10 million free tokens per day, including access to glm-5.3-flash. Log in through the latest CodeArts Agent IDE, JetBrains plugin, or VS Code plugin and the daily allowance is automatically credited. Tokens expire at midnight and do not roll over. This is not GLM API credit: the allowance can only be consumed inside Huawei Cloud CodeArts and cannot be exported to your own agent or third-party tools. The offer is limited-time and first-come, first-served; Huawei currently lists the end date as “to be announced.”