How do I use GLM 5.3 in Cline?
In Cline 4.1.25 or later, choose Standard Compute under API Provider (OpenAI Compatible on older versions), set Base URL to https://api.stdcmpt.com/v1, paste a Standard Compute key, enter z-ai/glm-5.3 as the Model ID and untick Supports Images: Cline runs on GLM 5.3 with no Z.ai account, on a plan from $19 a month that returns HTTP 402 to new requests once its compute budget is used.
# GLM 5.3. Untick Supports Images: it is text-only
# Cline 4.1.25+. On 4.1.24 and older, pick OpenAI Compatible instead.
API Provider Standard Compute
Base URL https://api.stdcmpt.com/v1
OpenAI Compatible API Key YOUR_API_KEY
Model ID z-ai/glm-5.3
# MODEL CONFIGURATION
Supports Images off (this model is text-only)
Context Window Size 400000
Max Output Tokens 32000
Reasoning Effort low
Cline keeps one provider and model per mode, so the Flash tab is either your Act model or a second profile you switch to.
Untick Supports Images
Cline ticks Supports Images for every Standard Compute or OpenAI Compatible model, so it sends screenshots, including the browser tool's, to GLM 5.3, which is text-only and answers with HTTP 400. Unticked, Cline stops sending them. For a task that needs eyes, switch the Model ID to z-ai/glm-5.3-flash, which reads images, with the box ticked.
GLM 5.3 or Flash for Cline
| GLM 5.3 | GLM 5.3 Flash | |
|---|---|---|
| Supports Images | Off | On |
| Reasoning Effort | Any level; none runs at low, since GLM always reasons | Same |
| Typical day at Z.ai list price | $3.70 | $0.41 |
| Busy day | $37 | $4.17 |
Z.ai list prices per million tokens: GLM 5.3 $1.40 input, $0.26 cached, $4.40 output; Flash $0.15, $0.03, $0.50. Source. A typical day is 8.6M tokens and a busy one 87.0M, 87% of input as cache reads: the median and 90th-percentile coding-agent day in our Agent Burn Index, measured on OpenCode, Cline, Kilo and similar agents, September 14 to October 13, 2026. Priced at the list rates above; your own days will differ.
Plan and Act on GLM
With "Use different models for Plan and Act modes" ticked, GLM 5.3 can plan and GLM 5.3 Flash can act; at list price Flash's day costs about 1/9 of GLM 5.3's. Cline warns that less capable models may not handle its complex prompts, so check Flash on a real task before trusting it with edits.
The catch
Two GLM-only routes cost less. Cline's built-in Z.AI Coding Plan provider (base https://api.z.ai/api/coding/paas/v4) takes a GLM Coding Plan key, which Z.ai sells from $18 a month, and ClinePass ($9.99) includes GLM 5.3 and 5.3 Flash as cline-pass/glm-5.3 and cline-pass/glm-5.3-flash.
Where Standard Compute fits
In Cline, changing vendor means changing one Model ID field; Base URL and key stay. GLM, Claude, GPT and Gemini share one published monthly budget.
FAQ
Should I use Cline's Z.AI provider or Standard Compute?
Cline's Z.AI and Z.AI Coding Plan providers need a Z.ai key. The Standard Compute provider, built into Cline since 4.1.25 (OpenAI Compatible on older versions), reaches GLM with a Standard Compute key, and the same key also serves other vendors' models.
Why does Cline compact my GLM session so early?
Context Window Size defaults to 128,000 for models Cline doesn't know. Raise it under Model Configuration; we suggest 400,000, since every extra token is resent on each turn.