GitHub issue #9651 in cline/cline, opened 2026-03-03 by NC17z, records a report that Cline v3.69.0 consistently constructs API requests that exceed the model's maximum context window even when the user message is extremely small, across MiniMax M2.5 and Qwen models on OpenRouter, because it serializes a roughly 4,555-token tool schema into every request, adds roughly 3,000-14,000 tokens of system prompt and text input, and requests the model's full context window as max_output_tokens, exceeding the maximum context by a small margin; the report says the behavior persists even after a full wipe and reinstall of VS Code and Cline, and gives a minimal reproduction of opening a brand-new empty folder and sending a simple message such as hi.
Verbatim Excerpts from Cited Records
“Cline is consistently constructing API requests that exceed the model’s maximum context window, even when the user message is extremely small (e.g., “hi”).”
·
cline/cline (issue 9651 opening body, What happened section)
·
2026-03-03T22:43:52Z
·GitHub Issue ↗
“This happens across multiple providers and models, including MiniMax M2.5 and Qwen models on OpenRouter.”
·
cline/cline (issue 9651 opening body, What happened section)
·
2026-03-03T22:43:52Z
·GitHub Issue ↗
“This behavior persists even after a full wipe and reinstall of VS Code and Cline.”
·
cline/cline (issue 9651 opening body, What happened section)
·
2026-03-03T22:43:52Z
·GitHub Issue ↗
“Send a simple message like: hi”
·
cline/cline (issue 9651 opening body, Steps to reproduce section)
·
2026-03-03T22:43:52Z
·GitHub Issue ↗
Counter-record: Issue closed 2026-11-13 by linear-code[bot], with no reason stated in the public thread; the linked Linear issue is CLINE-1477. 6 comments were read. rhenry74 (2026-03-05) reproduced the failure with MiniMax M2.5 on OpenRouter, requesting about 205,033 tokens against a 205,000-token maximum, and asked for a workaround. github-actions[bot] (2026-05-08) marked the issue stale after 60 days without activity. JBunga (2026-05-17) suggested shrinking the assembled prompt before it reaches the endpoint as a workaround test, not as a fix for the max_output_tokens calculation. gitmeatarru (2026-08-12) confirmed the same failure in Cline 4.0.12 with an Nvidia Nemotron model on OpenRouter, requesting 262,897 tokens against a 262,144-token maximum. The core report was never retracted or disproven in the thread, and the latest human report confirms the failure persisted into v4.0.12. No maintainer fix appears in the surveyed thread. The claim documents only the original report, not a verified current state.
Falsification probe: If GitHub issue #9651 is deleted without archive, if the author edits the opening body so the quoted sentences no longer match verbatim, or if the author posts a retraction of the report, this claim drops to Withdrawn.
Audit verification: Verified 100% continuous verbatim substring against official GitHub REST API by
scripts/verify_claim_verbatim.py.