From 61ab5c92722a636f535d12dcfb2180027ed75dcc Mon Sep 17 00:00:00 2001 From: chintondutta Date: Thu, 20 Aug 2026 12:11:02 +0530 Subject: [PATCH] docs: document GRAPHIFY_CLAUDE_CLI_MODEL in the env-var table claude-cli defaults to Opus, which is overkill for structured extraction and community labeling; the env var to override it existed in code but wasn't documented anywhere a user would find it before spending Opus quota on a job haiku handles fine (#2861). --- README.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/README.md b/README.md index 272c5f6f81..5a74dc398d 100644 --- a/README.md +++ b/README.md @@ -505,6 +505,7 @@ These are only needed for **headless / CI extraction** (`graphify extract`). Whe | `ANTHROPIC_API_KEY` | Claude (Anthropic) backend | `--backend claude` | | `ANTHROPIC_BASE_URL` | Anthropic-compatible endpoint URL (LiteLLM proxy, gateways, ...) | `--backend claude` (default: `https://api.anthropic.com`) | | `ANTHROPIC_MODEL` | Model name for the Claude backend — for custom endpoints, use the model name/alias your server exposes | `--backend claude` (default: `claude-sonnet-4-6`) | +| `GRAPHIFY_CLAUDE_CLI_MODEL` | Model override for the `claude-cli` backend, passed through as `claude -p --model` (accepts `haiku`, `sonnet`, or a full model ID) | optional — `--backend claude-cli` (default: Opus, the CLI's own default; overkill for structured extraction, and for `graphify label`/`cluster-only` specifically, `haiku` is usually enough) | | `GEMINI_API_KEY` or `GOOGLE_API_KEY` | Google Gemini backend | `--backend gemini` | | `OPENAI_API_KEY` | OpenAI or OpenAI-compatible APIs | `--backend openai` (local servers accept any non-empty value) | | `OPENAI_BASE_URL` | OpenAI-compatible server URL (llama.cpp, vLLM, LM Studio, ...) | `--backend openai` (default: `https://api.openai.com/v1`) | @@ -734,6 +735,7 @@ GRAPHIFY_OLLAMA_NUM_CTX=32768 graphify extract ./docs --backend ollama # overr GRAPHIFY_OLLAMA_KEEP_ALIVE=0 graphify extract ./docs --backend ollama # unload model after each chunk (saves VRAM on small GPUs) graphify extract ./docs --backend bedrock # AWS Bedrock via IAM - no API key, uses AWS credential chain graphify extract ./docs --backend claude-cli # route through Claude Code CLI - no API key, uses your Claude subscription +GRAPHIFY_CLAUDE_CLI_MODEL=haiku graphify cluster-only ./docs --backend claude-cli # cheaper model for community naming (claude-cli defaults to Opus) graphify extract ./docs --backend azure # Azure OpenAI (set AZURE_OPENAI_API_KEY + AZURE_OPENAI_ENDPOINT) graphify extract ./docs --max-workers 16 # AST parallelism (also GRAPHIFY_MAX_WORKERS) graphify extract --postgres "postgresql://user:pass@host/db" # introspect live PostgreSQL schema directly