Working with coding agents

Integrate Qorebit seamlessly with your favorite AI coding editors, extensions, and CLI frameworks. Enjoy unified billing and broad model availability in one place.

VS Code Extensions Integration

Connect your IDE plugins directly to Qorebit. Switch configuration modes using the tabs below:

Continue (YAML)

Continue (JSON)

Cline (Anthropic)

Cline (OpenAI)

Cursor

Configuration Settings

models:
  - name: Claude 3.5 Sonnet (Qorebit)
    provider: openai
    model: claude-3-5-sonnet
    apiBase: https://api.qorebit.ai/v1
    apiKey: qb_live_xxxxxxxxxxxxx

API Reference (Anthropic Compatible)

We provide a proxy path supporting the exact Anthropic messages structure, allowing you to use Anthropic SDKs unchanged.

POSThttps://api.qorebit.ai/v1/messages

Headers

Header	Required	Description
`x-api-key`	Yes	Your Qorebit API Key (starts with `qb_live_`)
`anthropic-version`	No	Anthropic API version protocol (e.g. `2023-06-01`)
`content-type`	Yes	Must be `application/json`

Request Body

JSON Request

{
  "model": "claude-sonnet-4-20250514",
  "max_tokens": 4096,
  "messages": [
    {
      "role": "user",
      "content": "Explain this codebase"
    }
  ]
}

Non-Streaming Response

JSON Response

{
  "id": "msg_abc123",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "This codebase is a FastAPI application..."
    }
  ],
  "model": "claude-sonnet-4-20250514",
  "stop_reason": "end_turn",
  "usage": {
    "input_tokens": 42,
    "output_tokens": 128
  }
}

Streaming Response

Trigger server-sent events by setting "stream": true in the body:

SSE Log Stream

event: message_start
data: {"type":"message_start","message":{"id":"msg_abc123","type":"message","role":"assistant","content":[],"model":"claude-sonnet-4-20250514","stop_reason":null,"usage":{"input_tokens":42,"output_tokens":0}}}

event: content_block_start
data: {"type":"content_block_start","index":0,"content_block":{"type":"text","text":""}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":"This"}}

event: content_block_delta
data: {"type":"content_block_delta","index":0,"delta":{"type":"text_delta","text":" codebase"}}

event: content_block_stop
data: {"type":"content_block_stop","index":0}

event: message_delta
data: {"type":"message_delta","delta":{"stop_reason":"end_turn"},"usage":{"output_tokens":128}}

event: message_stop
data: {"type":"message_stop"}

cURL Examples

Non-Streaming

Streaming

With Tools

cURL Request Options

curl -X POST https://api.qorebit.ai/v1/messages \
  -H "x-api-key: qb_live_xxxxxxxxxxxxx" \
  -H "anthropic-version: 2023-06-01" \
  -H "content-type: application/json" \
  -d '{
    "model": "claude-sonnet-4-20250514",
    "max_tokens": 1024,
    "messages": [
      {"role": "user", "content": "Hello, Claude!"}
    ]
  }'

SDK Examples

Python

Python Streaming

TypeScript

SDK Script Example

import anthropic

client = anthropic.Anthropic(
    api_key="qb_live_xxxxxxxxxxxxx",
    base_url="https://api.qorebit.ai",
)

message = client.messages.create(
    model="claude-sonnet-4-20250514",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Explain quantum computing"}
    ],
)

print(message.content[0].text)

Available Models

Model ID	Provider	Type
`claude-sonnet-4-20250514`	Anthropic	Premium
`claude-opus-4-20250514`	Anthropic	Premium
`claude-3.7-sonnet`	Anthropic	Premium
`claude-3.5-sonnet`	Anthropic	Premium
`gpt-4o`	OpenAI	Premium
`gpt-5`	OpenAI	Premium
`gemini-2.5-pro`	Google	Premium
`deepseek-v3`	DeepSeek	Open-weight
`deepseek-r1`	DeepSeek	Open-weight
`deepseek-v4-pro`	DeepSeek	Open-weight
`deepseek-v4-flash`	DeepSeek	Open-weight
`deepseek-prover-v2-671b`	DeepSeek	Open-weight (Math Reasoning)
`qwen-coder`	Alibaba	Open-weight
`qwen-3.5-397b`	Alibaba	Open-weight MoE (397B/17B)
`qwen-3.5-35b`	Alibaba	Open-weight MoE (35B/3B)
`qwen-3.5-27b`	Alibaba	Open-weight Dense (27B)
`qwen-3.5-2b`	Alibaba	Open-weight Dense (2B)

Using Open-Weight Models

Qorebit supports high-performance open-weight models natively. You can target these models in your favorite coding agents, IDE extensions, or CLI systems.

Running the Qwen 3.5 Flagship (397B MoE) Model

To run the high-performance Mixture-of-Experts Qwen 3.5 397B model, pass the model alias using the --model flag when launching Claude Code:

Bash / Zsh

# Launch Claude Code pointing to your Qorebit API Gateway
export ANTHROPIC_BASE_URL="https://api.qorebit.ai"
export ANTHROPIC_API_KEY="qb_live_xxxxxxxxxxxxx"

claude --model qwen-3.5-397b

Targeting within Cline & Roo Code Settings

To run open-weight models like the qwen-3.5-397b flagship within Cline or Roo Code settings:

Set the API Provider to Anthropic
Set the Base URL to https://api.qorebit.ai
Set the API Key to your Qorebit API Key (starts with qb_live_)
Select custom model and enter qwen-3.5-397b (or type it under custom models)

Rate Limits

Limit Type	Default Value
Requests per minute	60 requests / min
Requests per day	10,000 requests / day
Tokens per minute	200,000 tokens / min

Error Codes

HTTP Code	Error Type	Description
400	`invalid_request_error`	Malformed or unsupported request format
401	`authentication_error`	Missing or invalid Qorebit API key
403	`permission_error`	Your API key lacks access rights to this model
429	`rate_limit_error`	You have exceeded requests/tokens rate limits
500	`api_error`	Internal Qorebit gateway error
529	`overloaded_error`	Inference infrastructure is temporarily overloaded

Troubleshooting

"Invalid API key format"

Verify your API key starts with the prefix qb_live_. Regenerate your key from the Qorebit Settings dashboard if needed.

"Insufficient credits"

Top up your balance under Wallet → Top Up. Even open-weight models require credit balances to fund the GPU serving infrastructure.

Previous: Streaming Next: Claude Code