| Input length | Input / 1M | Output / 1M | Cache hit / 1M |
|---|---|---|---|
| 0 - 32k | ¥3.20 | ¥16 | ¥0.64 |
| 32k - 128k | ¥4.80 | ¥24 | ¥0.96 |
| 128k - 256k | ¥9.60 | ¥48 | ¥1.92 |
Doubao Seed 2.0 Code 在 NextModel 中是什麼?
香港團隊可用的 NextModel 目錄中的 Volcengine 模型,常用於 agentic coding、repository-scale refactors、long-context code review 工作負載。當前展示價格為輸入 ¥3.20 / 1M tokens、輸出 ¥16 / 1M tokens 每 100 萬 token,上下文視窗為 256k token。
適用場景
- agentic coding
- repository-scale refactors
- long-context code review
OpenAI 相容呼叫範例
保持 OpenAI SDK 呼叫方式不變,把 base_url 改為 NextModel,並使用模型目錄 ID doubao-seed-2-0-code。
from openai import OpenAI
client = OpenAI(
api_key="YOUR_API_KEY",
base_url="https://api.nextmodel.app/v1"
)
resp = client.chat.completions.create(
model="doubao-seed-2-0-code",
messages=[{"role": "user", "content": "Hello from NextModel"}]
)
print(resp.choices[0].message.content)相似替代項
Doubao Seed 2.0 Pro is Volcengine's flagship general-purpose model, onboarded through the Volcengine Agent Plan, sharing the same tiered pricing as the Code variant.
Qwen3 Coder Plus 是面向編碼的模型候選,適合需要中文工程工作流與長上下文程式碼輔助的團隊。
Qwen3 Max 是通用的高能力中文模型候選,適合在國內外提供方之間做比較的團隊。
常見問題
Doubao Seed 2.0 Code API 常見問題
What is Doubao Seed 2.0 Code best for?
Agentic coding tasks, large refactors, and long-context code review where the codebase exceeds a small context window.
How is Doubao Seed 2.0 Code priced?
Tiered by input length: ¥3.2/¥16 per 1M input/output tokens up to 32K tokens, rising to ¥9.6/¥48 above 128K tokens.