Doubao Seed 2.1 Pro API: Pricing, Context Window, Model ID and How to Call It

Short answer

Doubao Seed 2.1 Pro is the flagship of ByteDance's Doubao Seed 2.1 family, released on Volcano Engine's Ark platform in June 2026. On TeamoRouter its model ID is doubao-seed-2.1-pro. Volcano Engine's list price is ¥6 per million input tokens and ¥30 per million output tokens (about $0.86 and $4.29 at ¥7 to the dollar), with cached input at ¥1.20. The version Volcano Engine currently recommends, doubao-seed-2-1-pro-260915, has a 1024K-token context window and can return up to 256K tokens. On the day of writing (October 22, 2026) TeamoRouter charges about 76% of the official price; the live price is on the model page.

What Doubao Seed 2.1 Pro is

The Seed 2.1 family on Volcano Engine has three models. Official online-inference prices per million tokens:

Pro is the top and most expensive of the three. Volcano Engine's model list says it supports deep thinking, multimodal understanding, GUI tasks, tool calling and structured output. TeamoRouter currently offers Pro (doubao-seed-2.1-pro) and the previous-generation doubao-seed-2.0-lite.

Where it fits: Chinese-language work where you want a whole contract, long report or large chunk of code in one request; agent flows that call tools over several steps; backend endpoints that need the model to answer in a fixed JSON format.

Specs at a glance

From Volcano Engine's Ark model list:

Pricing

Volcano Engine list price (Ark pricing page, online inference, per million tokens):

Volcano Engine also sells a low-priority online tier and batch inference, both at ¥3 input and ¥15 output for Pro, for work that is not time-sensitive.

TeamoRouter bills per token at about 76% of the official price on the day of writing, with no monthly fee. Prices follow upstream changes; current numbers are on the model page and the pricing page.

A quick example: a request that reads 20,000 tokens and writes 2,000 costs ¥0.12 + ¥0.06 = ¥0.18 at the list price, and about 76% of that on TeamoRouter.

When TeamoRouter makes sense: you already use Claude, GPT or Gemini and want Doubao on the same key and balance, or you would rather not open a separate Volcano Engine account. If Doubao is all you need and you already have a Volcano Engine account, going direct is simpler.

Call it through TeamoRouter

Doubao works through the OpenAI-compatible endpoint; the base URL includes /v1. Create a key in the dashboard:

from openai import OpenAI

client = OpenAI(base_url="https://api.teamorouter.vip/v1", api_key="YOUR_TEAMOROUTER_KEY")
resp = client.chat.completions.create(
    model="doubao-seed-2.1-pro",
    messages=[{"role": "user", "content": "Turn these meeting notes into a to-do list: ..."}],
    max_tokens=8000,
)
print(resp.choices[0].message.content)

Or with curl:

curl https://api.teamorouter.vip/v1/chat/completions \
  -H "Authorization: Bearer YOUR_TEAMOROUTER_KEY" \
  -H "content-type: application/json" \
  -d '{
    "model": "doubao-seed-2.1-pro",
    "messages": [{"role": "user", "content": "Introduce yourself in three sentences."}]
  }'

Volcano Engine caps answers at 4K tokens by default, so set max_tokens higher for long output, as above. Full reference: API integration.

Use it in Cherry Studio

1. Open Settings → Model Services, click Add at the bottom of the provider list and choose OpenAI as the provider type.
2. Paste your TeamoRouter key and set the API URL to https://api.teamorouter.vip (no /v1; Cherry Studio adds it).
3. Click Fetch model list, select doubao-seed-2.1-pro and switch the provider on.

Step-by-step with screenshots: Cherry Studio setup.

Paying for usage

No subscription: you top up a balance from $5 and pay only for the tokens you use, by card or Alipay. The same balance covers Claude, GPT and Gemini.

FAQ

How does doubao-seed-2.1-pro relate to doubao-seed-2-1-pro-260915 on Volcano Engine? Volcano Engine model names carry a version date. On TeamoRouter you write doubao-seed-2.1-pro, without the date.

Why does the answer stop halfway? Volcano Engine's default answer limit is 4K tokens. Raise max_tokens in the request; the ceiling is 256K.

Can I use it in Codex? No. Codex sends requests through the Responses API, and TeamoRouter's /v1/responses supports GPT models only. For a desktop app, use a client that speaks the OpenAI-compatible API, such as Cherry Studio.

Is it cheaper than Qwen3.8-Max? At list prices, Doubao Seed 2.1 Pro is ¥6 input and ¥30 output; Qwen3.8-Max on Alibaba Cloud's China site is ¥12 and ¥36.

Do I need a Volcano Engine account? No. One TeamoRouter key covers Doubao, Claude, GPT and Gemini.

Next steps

• Doubao Seed 2.1 Pro model page: live price.
• Create an API key and send your first doubao-seed-2.1-pro request.
• Cherry Studio setup: use it without writing code.
• Qwen3.8-Max API: pricing and setup