Volcano Engine (Doubao) with OpenClaw

Volcano Engine (火山引擎) and BytePlus provide Doubao and other models. OpenClaw supports both the standard provider (volcengine) and the coding provider (volcengine-plan) for specialized coding models. BytePlus is the international variant with the same config pattern.

Onboard with the wizard:

openclaw onboard --auth-choice volcengine-api-key

Store the API key in credentials or set VOLCANO_ENGINE_API_KEY in your environment.

• volcengine — General models: Doubao, Kimi, GLM, DeepSeek
• volcengine-plan — Coding models (e.g. ark-code-latest, doubao-seed-code)

Both providers use the same API key. Config under models.providers.volcengine applies to both when credentials are shared.

• volcengine/doubao-seed-1-8-251228 — Doubao Seed 1.8
• volcengine/kimi-k2-5-260127 — Kimi K2.5
• volcengine/glm-4-7-251222 — GLM 4.7
• volcengine/deepseek-v3-2-251201 — DeepSeek V3.2
• volcengine-plan/ark-code-latest — Ark Coding

Check docs.openclaw.ai for current model IDs.

• Multi-model — Doubao, Kimi, GLM, DeepSeek via one provider
• Coding variants — volcengine-plan for code-focused models
• BytePlus — International variant, same setup

Volcano Engine exposes many models—use a capable model for user chat and a cheaper Doubao/GLM variant for cron and heartbeats. Patterns: example setups & model routing.

Volcano exposes Doubao and GLM IDs with date suffixes. After onboard, run openclaw models list and pin one flagship + one cheap model. APAC operators often pick Volcano for latency.

Account hygiene