[BUG] GLM-5 model pollutes context window with broken thinking output

Description

When using the GLM-5 model (Z.AI), OpenCode's context window gets polluted with broken/malformed thinking output. The terminal displays ? characters scattered throughout the interface, the rendered content becomes garbled, and the TUI layout breaks completely. The context window shows raw thinking tokens that should be hidden or properly parsed.

The GLM 4.7 model works correctly without any of these rendering issues under the same conditions.

Steps to Reproduce

Open OpenCode on Windows (cmd or PowerShell)
Select the GLM-5 (Z.AI Coding Plan) model
Run any task that triggers the model's thinking/reasoning mode (e.g., a build or code generation task)
Observe the terminal output becoming corrupted

Expected Behavior

The context window should display clean, properly formatted output — thinking tokens should be hidden or correctly rendered, just like the GLM 4.7 model does.

Actual Behavior

Multiple ? characters appear throughout the terminal (top, bottom, status bar)
The thinking/reasoning output leaks into the visible context window in a broken/malformed state
The TUI layout breaks — code snippets, markdown, and status indicators become garbled
The status bar at the bottom shows ???■■■■■ instead of proper indicators
The overall context window becomes unusable and filled with noise

Specific observations from the screenshot

The title bar and content area show ? characters on the left margin where they shouldn't be
Code content (Java — EvidenciaRunner.java) is partially rendered but interspersed with broken formatting
The bottom status bar shows: ? Build · glm-5 followed by more ? lines, then ? Build GLM-5 Z.AI Coding Plan
The very bottom shows ???■■■■■ esc interrupt — the progress/status indicators are corrupted

Comparison

Model	Behavior
GLM 4.7	Clean output, no artifacts, thinking is properly handled
GLM-5	Broken thinking tokens leak into context window, `?` characters everywhere, TUI breaks

Environment

OS: Windows 10/11
Terminal: cmd / PowerShell
OpenCode version: Latest
Model: GLM-5 (Z.AI Coding Plan)

Screenshot

I have verified that this issue has not already been requested
I have read the contributing guidelines

This issue might be a duplicate of existing issues. Please check:

Last token cut off in thinking (windows) #11194: Last token cut off in thinking (windows) — similar symptom of thinking tokens not rendering properly on Windows
Error "reasoning part reasoning-0 not found" from tool calls using GLM 4.5 Air #6486: Error "reasoning part reasoning-0 not found" from tool calls using GLM 4.5 Air — related GLM reasoning/thinking token handling issue

If your issue is distinct from these, feel free to keep it open and provide any additional context that differentiates it.

@erwinh22

Related issues and root cause

After investigating, this issue appears to be directly related to the known problem with <think>/<thinking> tag handling in OpenCode:

Support parsing <think> and <thinking> tags as reasoning blocks #11439 — Support parsing <think> and <thinking> tags as reasoning blocks (main tracking issue)
Thinking models: raw <think> tag content displayed instead of being collapsed #15380 — Thinking models: raw <think> tag content displayed instead of being collapsed (closed as duplicate of Support parsing <think> and <thinking> tags as reasoning blocks #11439)
thinking stop when use GLM-5 #15782 — thinking stop when use GLM-5
GLM 4.7 on Zai coding plan puts tool calls inside the thinking/reasoning tag. #6708 — GLM 4.7 puts tool calls inside thinking/reasoning tags
[bug] GLM 5 keeps screwing up the json parsing of read tool calling #13982 — GLM-5 keeps screwing up JSON parsing of read tool calling

Root cause

The extractReasoningMiddleware was intentionally removed in PR #11270 (v1.1.46+), which means <think> tags are now rendered as plain text instead of being collapsed into reasoning blocks. This causes the context window pollution seen in the screenshot — the GLM-5 thinking output leaks into the visible UI as garbled ? characters and broken formatting.

GLM 4.7 is unaffected because it handles reasoning differently and doesn't produce the same volume of raw <think> tag output.

Potential fix

PR #15382 by @erwinh22 implements the rendering-layer approach: tags are preserved in storage for multi-turn LLM context (respecting the decision in PR #11270), but parsed and rendered as reasoning blocks at display time in TUI, Web UI, and CLI. This PR appears to not have been merged yet.

This seems to still exist in GLM-5.1, and essentially makes opencode unusable.

Try setting the temperature to 1.0 in opencode.json:

	"provider": {
		"zai-coding-plan": {
			"models": {
				"glm-5": {
					"options": {
						"temperature": 1.0
					}
				},
				"glm-5.1": {
					"options": {
						"temperature": 1.0
					}
				}
			}
		}
	}

The transform.ts file doesn't have this config yet. With temperature 1.0, GLM-5 and GLM-5.1 feel more stable to me.

Try setting the temperature to 1.0 in opencode.json:

"provider": {
"zai-coding-plan": {
"models": {
"glm-5": {
"options": {
"temperature": 1.0
}
},
"glm-5.1": {
"options": {
"temperature": 1.0
}
}
}
}
}
The transform.ts file doesn't have this config yet. With temperature 1.0, GLM-5 and GLM-5.1 feel more stable to me.

its possible to set maximum context window? right now if above 80k is not useful, opencode still mark the context window is 200k so compaction not working and make thinking in a loop

Same for me - GLM 5.1 is also broken and unusable.

Actual outputs of GLM5.1 from opencode:

According to this Reddit post it seems also happening in Claude code.

So there seems to be two separate problems:

OpenCode parsing for GLM models are broken;
GLM models themselves are degrading, probably due to they've been overly quantized.

fix: render think tags as reasoning blocks + fix MCP MaxListeners ove…

Fixed by merging PR #15389, which renders “think” tags as reasoning blocks.
GLM-5’s thinking output will now be properly displayed in a reasoning block and won’t pollute the context.

fix: disable thinking/reasoning during compaction summarization

To stay organized issues are automatically closed after 60 days of no activity. If the issue is still relevant please open a new one.

Navigation Menu

Repository navigation

[BUG] GLM-5 model pollutes context window with broken thinking output #16903

Description

Steps to Reproduce

Expected Behavior

Actual Behavior

Specific observations from the screenshot

Comparison

Environment

Screenshot

Related issues and root cause

Root cause

Potential fix

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

[BUG] GLM-5 model pollutes context window with broken thinking output #16903

Description

Description

Steps to Reproduce

Expected Behavior

Actual Behavior

Specific observations from the screenshot

Comparison

Environment

Screenshot

Activity

github-actions commented on Mar 10, 2026

tayroneoliveira2023 commented on Mar 10, 2026

Related issues and root cause

Root cause

Potential fix

tareko commented on Mar 27, 2026

toninho09 commented on Apr 1, 2026

sa-altaiftech commented on Apr 3, 2026

jumski commented on Apr 6, 2026

Silur commented on Apr 8, 2026

ShibuyaCyana commented on Apr 8, 2026

saurav-shakya commented on Apr 10, 2026

github-actions commented on Jul 4, 2026

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions