Doubao LLM API API Pricing & Free Tier 2026
Verified May 14, 2026Doubao is ByteDance's self-developed LLM series, served via Volcengine Ark — Pro/Lite tiers covering dialog, long-context, vision, and embeddings.
Free tier
Doubao Pro / Lite tiers each include 500K free tokens valid 30 days on signup. Vision and image endpoints carry separate quotas.
Doubao Pro / Lite tiers each include 500K free tokens valid 30 days on signup. Vision and image endpoints carry separate quotas.
Per-model pricing
| Model | Context | Input $/1M | Output $/1M | Notes |
|---|---|---|---|---|
| Doubao-Pro-128k | 128K | $0.70 | $1.26 | Flagship long-context model |
| Doubao-Pro-32k | 32K | $0.11 | $0.28 | Mid-tier with 32K context |
| Doubao-Lite-32k | 32K | $0.04 | $0.08 | Low-cost tier for high-frequency tasks |
vs Global rivals
- Doubao-Pro-128k vs GPT-4o (openai)Doubao-Pro-128k is ~14% of GPT-4o pricing — the catalyst of the 2024 Chinese LLM price war.
- Doubao-Lite-32k vs Gemini 1.5 Flash (google)Comparable pricing; Doubao-Lite edges Chinese, Gemini 1.5 Flash edges English and multimodal.
Rate limits & access
Rate limit: Per-endpoint and per-model; enterprise verification scales limits.
API access requires Volcengine Ark "endpoint_id" creation. Mainland-China region default.
FAQ
- Doubao vs. Qwen — which to pick?
- Doubao is slightly cheaper; capability is comparable. Choose Doubao for ByteDance-ecosystem scenarios (Douyin etc.), Qwen for open-source breadth and multilingual reach.
- Why is an endpoint_id required before calling?
- Volcengine Ark abstracts model versions behind endpoints, enabling gradual rollout — an externalization of ByteDance's internal ML-engineering practice, making multi-model multi-version management cleaner.
Prices in USD. Some providers (DashScope, Doubao) bill in CNY; figures shown at ~7.2 CNY/USD reference rate. Verify on the provider's official page before procurement.