Doubao API
ByteDance's Seed (Doubao) models through BytePlus ModelArk internationally and Volcano Engine Ark in mainland China, with an OpenAI-compatible API.
9 open-source projects declare Doubao API in their code; no maker's live product we track uses it yet.
Profile checked
- Oct 2026 Now covered: Doubao API, for llm api.
How makers use Doubao API
What Doubao API is used for
Doubao API vs the alternatives makers chose
| Option | Makers' products | 2026 launches | Free for | At 10,000 users | Open source | |
|---|---|---|---|---|---|---|
| 736 80% | 77% | ~1,000 users | $10/mo · GPT-6 Luna | No | ||
| 252 27% | 21% | ~1,000 users | $10/mo · Claude Haiku 5.5 | No | ||
| 99 11% | 12% | ~360 users | $28/mo · Gemini 3.1 Flash-Lite | No | ||
| 39 4% | 3% | ~1,500 users | $6.75/mo · GPT OSS 20B | No | ||
| 22 2% | 0% | — | — | No | ||
| 15 2% | 1% | ~1,700 users | $6.00/mo · Ministral 3 (3B) | No | ||
| Doubao API | 0 0% | 0% | ~1,100 users | — | No |
For LLM API: how many of the 923 makers' products we know the LLM API of chose each option, and the share among the 256 launched in 2026, since a tool that came first has products from every year, and what each costs at 10,000 users.
Costs are the cheapest plan that fits, from each vendor's pricing page, at the usage our stack planner assumes for a product with that many users. Free for: the users its cheapest plan a business can use stays free for.
When to use Doubao API, and when not
Something else fits better if
- You want the strongest general models and the widest ecosystem of examples and integrations: OpenAI API or Claude or Gemini API.
- Token cost dominates your budget, for example high-volume batch or agent workloads: DeepSeek or Groq or Mistral AI.
- You want to try or mix many models with one key and one bill, and fall back when a provider is down: OpenRouter.
- You want that routing and spend control but self-hosted, with keys and logs on your servers: LiteLLM.
ByteDance sells API access to its own Seed models, called Doubao in China, through two platforms: BytePlus ModelArk for international customers and Volcano Engine Ark for mainland China. ModelArk serves the Seed 2.0 Pro, Lite and Mini models and Seed 2.1 Turbo with up to 256K tokens of context, alongside ByteDance's Seedream image and Seedance video models and resold third-party models such as DeepSeek and GLM. Some newer models, such as Doubao Seed 2.1 Pro and Lite, are only on the mainland platform.
More about Doubao API
ModelArk's chat API is compatible with the OpenAI API: point the OpenAI SDK at https://ark.ap-southeast.bytepluses.com/api/v3 with a ModelArk API key. BytePlus also publishes its own SDKs for Python (arkruntime), Go and Java, and lists Responses and Messages APIs, batch inference, context caching and a Managed Agents API in the same reference.
Useful for a small team: each model comes with a free trial quota, and a "Free Tokens Only" mode suspends calls before you'd be billed; batch and "flex" online inference run at roughly half the standard price; and you can turn on application-layer encryption so requests are only decrypted inside ModelArk's inference sandbox.
Limits: prices step up for longer prompts, and context caching adds an hourly storage fee. ByteDance is headquartered in Beijing; for ModelArk, BytePlus says customer data is processed in Malaysia, Indonesia and/or the EU/EEA and isn't used for its own model training without your authorization, but inputs and outputs caught by the content filter are kept for 180 days in Malaysia.
Who uses it
No maker's live product on record yet, and 9 open-source projects that declare it in their code.
We haven't found a maker's live product that uses Doubao API yet — only the open-source projects below, which declare it in their code.
None of these in makers' own words.
Open source: a project that declares Doubao API as a dependency in its public code — verifiable, but not necessarily a live product.
Alternatives to Doubao API
All alternatives by situation →Further reading
Questions makers ask about Doubao API
What are the alternatives to Doubao API?
For LLM API, the other options makers' products we track chose most are OpenAI API (736), Claude (252), Gemini API (99), Groq (39), Azure OpenAI (22), Mistral AI (15). Each alternative, with the situation it fits, is on the Doubao API alternatives page.
Where is my data processed?
BytePlus says ModelArk may use data centers in Malaysia, Indonesia and/or the EU/EEA for model deployment and customer data processing, with a global load balancer routing traffic among them. source ↗
Does BytePlus train on my data?
No, not without your prior authorization. BytePlus describes itself as a data processor that handles customer data only on your instructions and keeps each customer's data isolated from others. source ↗
Are prompts and outputs retained?
Content that triggers the content filter, both input and output, is kept for 180 days in Malaysia for content-safety purposes. source ↗
Can I use the OpenAI SDK?
Yes. Set the base URL to https://ark.ap-southeast.bytepluses.com/api/v3 and authenticate with a ModelArk API key as a bearer token. Coding Plan subscribers use a different base URL. source ↗
Is there a free tier?
A trial quota. New BytePlus users get free inference tokens per model, shared across the main account and its fine-tuned versions, and Free Tokens Only mode suspends calls once the 500K-token quota is nearly used. After that you activate each model for pay-as-you-go. source ↗
Which SDKs are official?
BytePlus recommends its ModelArk SDKs for Python (arkruntime), Go and Java, keeps the older byteplus-python-sdk-v2 for APIs the new SDK doesn't cover yet, and supports community OpenAI-compatible SDKs. source ↗
Is Doubao API free?
No permanent free tier; paid use starts at Pay per token. source ↗
Is Doubao API open source or self-hostable?
Not open source, and hosted only.
Can AI coding agents work with Doubao API?
No llms.txt, official MCP server or CLI found yet.
Who uses Doubao API?
No maker's live product we track yet; 9 open-source projects declare it in their code. source ↗