Answer: Models added to GitHub Copilot CLI must support tool calling (also called function calling) and streaming. Token limits should be configured to match the model, and for an unknown model Copilot CLI can default to a limit of 128,000 prompt tokens. Status: Confirmed Entry: tool/github-copilot/models/model-requirements Revision: 3 Digest: sha256:c414095e0958b84e (changes whenever the content changes) Timeline: Created: 2026-11-13T03:23:56Z Last changed: 2026-11-20T13:02:10Z (Default prompt token limit for unknown model: — → 128,000) Last verified: 2026-11-20T13:02:10Z Next check: 2026-11-21T13:02:10Z Check interval: Daily Expected change: No fixed schedule Data: Field | Value | Unit | Source | As of Required capability 1 | tool calling | | [1] | 2026-11-20 Required capability 2 | streaming | | [1] | 2026-11-20 Default prompt token limit for unknown model | 128,000 | prompt tokens | [1] | 2026-11-20 Sources: [1] GitHub. "Adding LLM models to GitHub Copilot CLI - GitHub Docs". https://docs.github.com/en/copilot/how-tos/copilot-cli/customize-copilot/add-llm-models Role: primary. Accessed: 2026-11-20T13:02:10Z Quote: "Choose a model and endpoint that support tool calling (also called function calling) and streaming" (Supports: Required capability 1, Required capability 2) Quote: "Copilot CLI can default to a limit of 128,000 prompt tokens" (Supports: Default prompt token limit for unknown model) Snapshot SHA-256: 374f3e172a965fc1265573badeadebe788c463cf9524f40e64412c0a46488827 (captured 2026-11-19T12:55:54Z) Verification: Method: Automated extraction from [1] with deepseek-flash (JSON output, thinking disabled). Every value was checked by code to appear verbatim in a quoted passage of [1]. On this check the text of [1] was unchanged since 2026-11-19T12:55:56Z, so the values verified then were kept. Cross-check: not-available at 2026-11-20T13:02:10Z. No independent second source is configured for this entry Last successful check: 2026-11-20T13:02:10Z Last attempt: 2026-11-20T13:02:10Z Notes and limitations: Extracted automatically from a single official source (section "Model requirements"). Related: https://latestn.com/tool/github-copilot/models/models-by-plan Which AI models are available with each GitHub Copilot plan? https://latestn.com/tool/github-copilot/models/models-per-client Which AI models are supported in GitHub Copilot per client? https://latestn.com/tool/github-copilot/models/models-per-plan Which AI models are supported in GitHub Copilot for each Copilot plan? https://latestn.com/tool/github-copilot/models/manual-provider-config How do I manually add a model provider configuration to GitHub Copilot CLI? https://latestn.com/tool/github-copilot/models/model-availability-factors What determines which models are available to me in GitHub Copilot? https://latestn.com/tool/github-copilot/models/model-hosting-providers Which AI model providers host the models available in GitHub Copilot? https://latestn.com/tool/ All entries in AI tools Change log: Time | Event | Change | Source 2026-11-20T13:02:10Z | changed | Default prompt token limit for unknown model: — → 128,000 | [1] 2026-11-20T13:02:10Z | changed | Required capability 2: — → streaming | [1] 2026-11-20T13:02:10Z | changed | Required capability 1: — → tool calling | [1] 2026-11-20T13:02:10Z | changed | default prompt token limit for unknown model: 128,000 → — | [1] 2026-11-20T13:02:10Z | changed | streaming requirement: streaming → — | [1] 2026-11-20T13:02:10Z | changed | tool calling requirement: tool calling → — | [1] 2026-11-17T13:18:17Z | changed | streaming requirement: — → streaming | [1] 2026-11-17T13:18:17Z | changed | tool calling requirement: — → tool calling | [1] 2026-11-17T13:18:17Z | changed | required capability 2: streaming → — | [1] 2026-11-17T13:18:17Z | changed | required capability 1: tool calling → — | [1] 2026-11-13T03:23:56Z | added | Entry created | — Changes for this entry as a feed: https://latestn.com/tool/github-copilot/models/model-requirements/latest Challenge this entry: Open challenges: None How: If a value here is wrong or out of date, send what is wrong and a link to evidence. Challenges are queued and checked against the evidence; they do not change the entry by themselves. If the entry changes as a result, the change log says so. Form: https://latestn.com/dispute/tool/github-copilot/models/model-requirements POST: https://latestn.com/dispute/tool/github-copilot/models/model-requirements with claim, evidence_url, and optionally field and contact (form-encoded or JSON) GET: https://latestn.com/dispute/tool/github-copilot/models/model-requirements?claim={what is wrong}&evidence_url={link} How to cite: LatestN. (2026). What are the model requirements for adding LLM models to GitHub Copilot CLI? Revision 3, last verified 2026-11-20T13:02:10Z. https://latestn.com/tool/github-copilot/models/model-requirements (accessed 2026-11-21T09:52:17Z). @misc{LatestN_tool_github_copilot_models_model_requirements_r3, title = {What are the model requirements for adding LLM models to GitHub Copilot CLI?}, author = {{LatestN}}, year = {2026}, howpublished = {\url{https://latestn.com/tool/github-copilot/models/model-requirements}}, note = {Revision 3; digest sha256:c414095e0958b84e; last verified 2026-11-20T13:02:10Z; accessed 2026-11-21T09:52:17Z} } Reuse: License: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/) Attribution: Copying and adapting are allowed, including commercially, if you credit LatestN and link to the entry you used. Each entry ends with a ready-made citation. URL: https://latestn.com/tool/github-copilot/models/model-requirements