BytePlus

💰 Plan & Pricing

✅

Free Tier — 500K Tokens/LLM, 2M/Vision, 200 Images
Each LLM model gets 500K free tokens. Vision models get 2M tokens. 200 free 4K image generations via Seedream. No credit card required to start.

✅

ArkCode Pro — $9/mo Coding Plan (Verified ✅)
Full access to all 9 Coding Plan models including dola-seed-2.0-lite (84 tok/s), dola-seed-2.0-pro (deep-thinking), and ark-code-latest (intelligent routing). Anthropic-compatible endpoint for Claude Code, Cursor, Hermes Agent, and other IDEs.

🔑 API Key

ark-add3...297be

🌐 Endpoints

https://ark.ap-southeast.bytepluses.com/api/coding/v3

Coding Plan (OpenAI-compatible) — use this for Coding Plan quota

https://ark.ap-southeast.bytepluses.com/api/coding

Coding Plan (Anthropic-compatible) — for Claude Code, Hermes Agent, etc.

https://ark.ap-southeast.bytepluses.com/api/v3/

General API — ⚠️ requires ep-xxx endpoint, incurs additional charges

📦 Coding Plan Models $9/mo · All ✅ Verified

Model	Speed (tok/s)	Type	Notes
dola-seed-2.0-lite	⚡ 84 tok/s	Coding	⚡ Fastest
dola-seed-2.0-code	⚡ 80 tok/s	Coding	IDE-optimized
dola-seed-2.0-pro	⚡ 50 tok/s	Reasoning	Deep-thinking multimodal
glm-4.7	⚡ 71 tok/s	Chat	GLM 4.7
bytedance-seed-code	⚡ 63 tok/s	Coding	ByteDance Seed
glm-5.1	⚡ 30 tok/s	Chat	GLM 5.1
kimi-k2.5	⚡ 26 tok/s	Chat	Moonshot Kimi
ark-code-latest	⚡ 15 tok/s	Auto	Intelligent routing
gpt-oss-120b	⚡ 15 tok/s	Chat	Open-source GPT

🌐 General API Models Requires ep-xxx Endpoint

Model	Type	Context	Notes
Doubao-Pro-32K	LLM	32K	Doubao standard
Doubao-Pro-128K	LLM	128K	Doubao long context
Doubao-Lite-32K	LLM	32K	Fast & cheap
Doubao-Lite-128K	LLM	128K	Fast cheap long context
Dola-Seed-2.0-pro	Multimodal	—	Deep-thinking
Dola-Seed-2.0-Code	Coding	—	IDE-agent optimized
DeepSeek-V3.2	LLM	—	Reasoning

General API models require dedicated inference endpoints (ep-xxx). Create them in the BytePlus console under ModelArk.

🎨 Image & Video Models

Model	Type	Notes
Dola-Seedream-5.0-lite	Image Gen	4K output, text+image input, batch generation
Dreamina-Seedance-2.0	Video Gen	Multi-modal video generation
Dreamina-Seedance-2.0-fast	Video Gen	Fast video generation

🔬 Other Models

Model	Type	Notes
OmniHuman	Digital Human	Film-grade from photo + audio + text
Seed Speech	TTS	Large model voice synthesis

🔧 Configuration — Claude Code / Hermes IDE

Use these environment variables for Anthropic-compatible IDEs (Claude Code, Cursor, Hermes Agent, etc.)

ANTHROPIC_AUTH_TOKEN=ark-add3...297be
ANTHROPIC_BASE_URL=https://ark.ap-southeast.bytepluses.com/api/coding
ANTHROPIC_MODEL=ark-code-latest

Supported IDEs: Claude Code, Cursor, OpenCode, Cline, Roo Code, TRAE, Kilo Code, Hermes Agent

🔧 Configuration — OpenAI Style

For tools using the OpenAI SDK format

base_url=https://ark.ap-southeast.bytepluses.com/api/coding/v3
api_key=ark-add3...297be

💻 cURL Example

curl -X POST https://ark.ap-southeast.bytepluses.com/api/coding/v3/chat/completions \
  -H "Authorization: Bearer ***" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "ark-code-latest",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

🐍 Python Example

from openai import OpenAI

client = OpenAI(
    api_key="ark-add3...297be",
    base_url="https://ark.ap-southeast.bytepluses.com/api/coding/v3"
)

response = client.chat.completions.create(
    model="ark-code-latest",
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)

🚨 Error Reference

Error	Meaning	Fix
401 — key doesn't exist	Wrong domain	Use `bytepluses.com` — NOT `volces.com`
400 — InvalidSubscription	Coding Plan not activated	Subscribe to ArkCode Pro ($9/mo) at Volcengine console
404 — InvalidEndpointOrModel	Wrong endpoint for model type	General API models need an `ep-xxx` endpoint, not `/api/coding`

🖥️ Console & Management

Volcengine Coding Plan:
https://console.volcengine.com/ark/region:ark+cn-beijing/openManagement?OpenModelVisible=false&tab=CodingPlan

BytePlus Console:
https://console.byteplus.com → ModelArk

Use Volcengine console to activate/manage Coding Plan subscription. Use BytePlus console for General API endpoints and model management.

⚠️ Pitfalls & Notes

🚨

CRITICAL: Use bytepluses.com domain! — Do NOT use volces.com — it returns 401 for the same key. Always use bytepluses.com.

🚨

Don't use /api/v3 with Coding Plan key! — Using the General API endpoint /api/v3 with a Coding Plan key incurs ADDITIONAL charges, not plan quota. Always use /api/coding/v3 for Coding Plan.

✅

Coding Plan Verified & Working — All 9 Coding Plan models tested and confirmed operational (May 2026). The $9/mo ArkCode Pro subscription is active and functional.

ℹ️

Coding Plan Supported IDEs — Claude Code, Cursor, OpenCode, Cline, Roo Code, TRAE, Kilo Code, Hermes Agent

ℹ️

Dola-Seed-2.0-Code — A unique coding-optimized model designed specifically for IDE workflows.

ℹ️

General API Endpoints — Doubao, DeepSeek, and other General API models require creating dedicated ep-xxx inference endpoints in the console before use.

🏷️ Categories

Chat Coding Vision Reasoning Image Gen Video Gen TTS

📱 BytePlus