Gemini CLI error
Gemini CLI: API Error: Model stream ended with an invalid chunk or missing finish reason
[API Error: Model stream ended with an invalid chunk or missing finish reason.]Status: WorkaroundLast verified 1 documented fix4 sources
What it means
The model's streamed response ended without a valid finish reason or content, so the turn stopped mid-task. Current source raises this as InvalidStreamError with types such as NO_FINISH_REASON and MALFORMED_FUNCTION_CALL (for example "Model stream ended without a finish reason.").
Causes
- The response stream from the model ended without a finish reason or with malformed content (InvalidStreamError in core/geminiChat.ts). No root cause beyond this was stated by maintainers in #7851.
How to fix it
- Fix 1.
Update the CLI. In #7851 a maintainer (SandyTao520) asked the reporter to update to v0.3.4. Invalid-stream handling was later changed in v0.7.0 (#9169 "Retry invalid stream only 1 time") and v0.42.0 (#26218 "handle InvalidStream event gracefully without throwing").
Terminalnpm install -g @google/gemini-cli@latest
More Gemini CLI errors
All 19 Gemini CLI errorsYou must be a named user on your organization's Gemini Code Assist Standard edition subscription to use this service. Please contact your administrator to request an entitlement to Gemini Code Assist Standard edition.Status: WorkaroundFailed to sign in. Message: Your current account is not eligible... because it is not currently available in your location.Status: By designFailed to sign in. Message: Request contains an invalid argumentStatus: WorkaroundUNABLE_TO_GET_ISSUER_CERT_LOCALLY or unable to get local issuer certificateStatus: WorkaroundThis account requires setting the GOOGLE_CLOUD_PROJECT or GOOGLE_CLOUD_PROJECT_ID env var. See https://geminicli.com/docs/get-started/authentication/#set-gcpStatus: By designInvalid Google Cloud Project ID: "<projectId>". The GOOGLE_CLOUD_PROJECT (or GOOGLE_CLOUD_PROJECT_ID) environment variable must be set to your string-based Project ID (e.g., "my-project-123"), not your numeric Project Number. Please update your environment variables.Status: By designFailed to login. Message: Precondition check failed.Status: Fixed in releaseYou are currently configured to use a Google Cloud Project but lack a Gemini Code Assist license. Please contact your administrator to request a license.Status: Workaround
Related reading
- Do Your Agent Hooks Fail Open? Claude Code vs Codex vs Cursor vs Gemini CLIA guard hook that crashes, times out or points at a missing script lets the tool call through by default in Claude Code. v2.1.295 adds onFailure: block. Here is what each coding agent does when your policy hook breaks, with real test runs.
- How to Keep API Keys Away From Your AI Coding Agent: Claude Code vs Codex vs Cursor vs Gemini CLIA deny rule on .env stops cat but not a Python one-liner. Here are the three layers that actually keep secrets out of a coding agent's context - file rules, environment scrubbing and credential masking - and what Claude Code, Codex, Cursor and Gemini CLI give you for each, with real test runs.
- Headless AI Coding Agents in CI: Claude Code, Codex CLI, Gemini CLI, and opencode ComparedHeadless AI coding agents in CI compared: Claude Code, Codex CLI, Gemini CLI, and opencode on invocation flags, sandboxing, auth, and output formats.
- Gemini CLI to Antigravity CLI Migration Guide: The June 18 DeadlineGemini CLI to Antigravity CLI migration guide: install, migrate config, update scripts and CI, and avoid the silent MCP failure that breaks tool calls.
- Gemini CLI Guide: 1M Token Context and Who Can Still Use ItGemini CLI puts a 1 million token context in your terminal but now only serves paid Gemini API, Vertex AI and Code Assist users. Setup and TypeScript workflow.
FAQ
What does "API Error: Model stream ended with an invalid chunk or missing finish reason" mean in Gemini CLI?
The model's streamed response ended without a valid finish reason or content, so the turn stopped mid-task. Current source raises this as InvalidStreamError with types such as NO_FINISH_REASON and MALFORMED_FUNCTION_CALL (for example "Model stream ended without a finish reason.").
How do I fix "API Error: Model stream ended with an invalid chunk or missing finish reason"?
Update the CLI. In #7851 a maintainer (SandyTao520) asked the reporter to update to v0.3.4. Invalid-stream handling was later changed in v0.7.0 (#9169 "Retry invalid stream only 1 time") and v0.42.0 (#26218 "handle InvalidStream event gracefully without throwing").
Is this Gemini CLI error fixed?
Status: Workaround. No release removes the cause. The documented fixes work around it. Affected versions: Reported on 0.3.3 (#7851)
Sources
Last verified against 4 sources: official docs, maintainer replies and release notes.
Newsletter
Fixes for the tools you use, weekly
One email per week. New coding agent releases, what broke, and how developers are working around it.