Support
For payments customer support, our top pick is GLM-5.3 Flash from Z.ai. Low cost at 18B active parameters with strong tool use for account lookups. DeepSeek V4.1 Flash is a close second, and Qwen3.8 2.4T-A95B rounds out the list.
Support agents answer merchants and cardholders about declines, payouts, and refunds using real account data, and deflect escalations from engineering. They need to be fast, inexpensive at volume, and good in many languages.
Updated
Low cost at 18B active parameters with strong tool use for account lookups.
A natively multimodal GLM that Z.ai says beats GLM-5.2 at one-tenth the price.
Input: text, image
Harnesses in Runtime
Very fast responses for live chat and SMS.
A multimodal MoE that activates only 8B parameters per token and shrinks the KV cache about 4x.
Input: text, image
Harnesses in Runtime
Broad multilingual coverage for global merchant bases.
The first Qwen-Max-class model released with open weights.
Input: text
Harnesses in Runtime
Turn the steps your team already follows into a skill the agent reads before every run.
Run two or three of these models against past cases. Evals show which one passes at the lowest cost.
The agent works on its own computer in your VPC, with role-based access and an approval before sensitive actions.
For payments customer support, our top pick is GLM-5.3 Flash from Z.ai. Low cost at 18B active parameters with strong tool use for account lookups. DeepSeek V4.1 Flash is a close second, and Qwen3.8 2.4T-A95B rounds out the list.
Fast and inexpensive at high volume. Multilingual answers. Good tool use for account lookups. Test two or three candidates against your own past cases before you commit, and keep a human approval on any step that moves money or changes a customer's status.
Yes. Download the weights and serve them with vLLM or SGLang in your own cloud, or use Amazon Bedrock or Google Vertex AI in your account. With Runtime, the agent's computer also runs in your VPC, so card data and PII stay in your environment.
GLM-5.3 Flash (MIT), DeepSeek V4.1 Flash (MIT) are permissive and fine for commercial use. Qwen3.8 2.4T-A95B (Qwen3.8-Max License) uses a custom license, so read the terms first.
