Claude Code

Claude Code is Anthropic's terminal-based coding agent. It speaks only the Anthropic Messages API — it cannot talk to an OpenAI-style Chat Completions endpoint directly.

Recommended: honey-llm-proxy

honey-llm-proxy is GreenPT's own translation proxy — a single small Rust binary (no Python, no LiteLLM) that lets both Claude Code and Codex use GreenPT models. It translates the Anthropic Messages API to Chat Completions, including streaming and tool calling.

Install & run

cargo install --path .
export GREENPT_API_KEY="your-greenpt-key"
greenpt-proxy            # listens on http://127.0.0.1:4000

Options: --port N (or PORT env), and GREENPT_BASE_URL (default https://api.greenpt.ai/v1).

Point Claude Code at the proxy

export ANTHROPIC_BASE_URL="http://localhost:4000"
export ANTHROPIC_AUTH_TOKEN="dummy"
export ANTHROPIC_DEFAULT_OPUS_MODEL="glm-5.2"
export ANTHROPIC_DEFAULT_SONNET_MODEL="glm-5.2"
export ANTHROPIC_DEFAULT_HAIKU_MODEL="minimax-m2.5"
claude

The ANTHROPIC_DEFAULT_*_MODEL variables map Claude Code's internal model tiers (opus / sonnet / haiku) to GreenPT model IDs — here glm-5.2 handles the heavy work and minimax-m2.5 (the cheapest model) covers the fast/background tier Claude Code uses for lightweight internal tasks. The proxy passes model names through verbatim, so any GreenPT ID works (glm-5.2, minimax-m2.5, kimi-k2.6).

To make it permanent instead of exporting each session, add these to ~/.claude/settings.json under "env", or wrap them in a shell function:

greenpt-code() {
  export ANTHROPIC_BASE_URL="http://localhost:4000"
  export ANTHROPIC_AUTH_TOKEN="dummy"
  export ANTHROPIC_DEFAULT_OPUS_MODEL="glm-5.2"
  export ANTHROPIC_DEFAULT_SONNET_MODEL="glm-5.2"
  export ANTHROPIC_DEFAULT_HAIKU_MODEL="minimax-m2.5"
  claude "$@"
}

Then greenpt-code starts a GreenPT-backed session while plain claude keeps using your regular Anthropic account.

Verify

curl -s localhost:4000/v1/messages -H 'content-type: application/json' \
  -d '{"model":"glm-5.2","max_tokens":50,"messages":[{"role":"user","content":"Say hi"}]}'

A successful Anthropic-format reply confirms the proxy is translating correctly.

Alternative: LiteLLM or claude-code-router

Any Anthropic→Chat-Completions translator works. LiteLLM in proxy mode is the most common general-purpose option:

# litellm-config.yaml
model_list:
  - model_name: glm-5.2
    litellm_params:
      model: openai/glm-5.2
      api_base: https://api.greenpt.ai/v1
      api_key: os.environ/GREENPT_API_KEY

pip install 'litellm[proxy]'
litellm --config litellm-config.yaml --port 4000
export ANTHROPIC_BASE_URL="http://localhost:4000"
export ANTHROPIC_AUTH_TOKEN="dummy"
export ANTHROPIC_DEFAULT_SONNET_MODEL="glm-5.2"
claude

claude-code-router follows the same architecture. These general-purpose proxies add a local hop and occasional protocol-translation rough edges (especially around tool calling and usage reporting) — honey-llm-proxy is purpose-built for GreenPT and handles both Claude Code and Codex.

Troubleshooting