ByteDance Doubao · Rate Limits

Doubao Rate Limits

Volcano Engine enforces per-endpoint RPM/TPM and concurrent quotas, configurable per workspace/endpoint. Limits visible in the Ark console.

3 Limits Throttle: 429
AILLMByteDanceRate Limiting

Limits

Per-Endpoint RPM endpoint
requests-per-minute
see Ark console
Configurable per deployed model/endpoint.
Per-Endpoint TPM endpoint
tokens-per-minute
see Ark console
Concurrency endpoint
concurrent-requests
see Ark console

Policies

Backoff Strategy
Exponential backoff with jitter; honor Retry-After.
Reserved Capacity
Reserved-instance subscriptions guarantee throughput beyond shared quotas.

Sources