GLM-5.1 Metered Chat (Large Context)

Choose this endpoint when you need a reasoning-tier model (GLM-5.1-FP8) with large-context input support and want metered, pay-per-call pricing without a subscription — ideal for bursty workloads, agent orchestration steps, or long prompts that exceed smaller-context route limits. Prefer over the flat-rate GLM-5.1 sibling when your input is large and you want metered billing. Prefer over other Telnyx models (Gemma, Llama, DeepSeek) when you specifically need GLM-5.1's reasoning capability rather than a smaller or faster model.

• Payment failure: insufficient USDC balance or x402 payment rejected — returns 402 Payment Required
• Input too long: messages array exceeds 32 items or content exceeds 200,000 characters — returns 400 validation error
• Invalid model: any model string other than zai-org/GLM-5.1-FP8 is rejected
• max_tokens out of range: values above 8192 or below 1 return a 400 error
• Temperature out of range: values outside 0-2 return a validation error
• More than 16 tools provided: returns a 400 error
• More than 1 completion requested (n>1): rejected as n is fixed to 1