PcPc.AI Developer API

AI API Documentation

Gateway

Gateway URL

https://ai.pcpc.ai

OpenAI Base URL

https://ai.pcpc.ai/v1

Anthropic Base URL

https://ai.pcpc.ai/anthropic

Models

Prices are in USD per million tokens; tool support is shown by the icons.

Model prices are never charged in real dollars — they only meter your quota. Buy a membership plan and a USD API allowance is granted every week.

Model	Tools	Context	Input (cache hit) / output price	Notes
DeepSeek V4 Flash 0731 deepseek-v4-flash	—	1M	$0.22($0.022)/$0.66	50% of the official price.
DeepSeek V4 Pro 0813 deepseek-v4-pro	—	1M	$0.66($0.066)/$1.98	50% of the official price.
Gemini 3.7 Flash gemini-3.7-flash	—	1M	$0.56($0.056)/$2.81	75% of the official price.
GLM 5.3 glm-5.3		1M	$0.49($0.091)/$1.54	—
Kimi K3 Kimi-K3	—	1M	$1.40($0.14)/$7.00	50% of the official price.
Claude Opus 5 claude-opus-5	—	1M	$2.50($0.25)/$12.50	50% of the official price.
Grok 4.6 grok-4.6		1M	$1.00($0.10)/$3.00	50% off
GPT 5.6 Sol gpt-5.6-sol		1M	$2.00($0.20)/$10.00	50% off
GPT 5.6 Luna gpt-5.6-luna		1M	$0.20($0.02)/$1.20	—
GPT 5.6 Terra gpt-5.6-terra		1M	$1.00($0.10)/$6.00	50% off

API Examples

OpenAI Chat Completions

curl https://ai.pcpc.ai/v1/chat/completions \
  -H "Authorization: Bearer $PCPC_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4-flash",
    "messages": [{"role": "user", "content": "Hello"}],
    "stream": true
  }'

OpenAI Responses

curl https://ai.pcpc.ai/v1/responses \
  -H "Authorization: Bearer $PCPC_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4-flash",
    "input": "Hello",
    "stream": true
  }'

Anthropic Messages

curl https://ai.pcpc.ai/anthropic/v1/messages \
  -H "x-api-key: $PCPC_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-v4-flash",
    "max_tokens": 1024,
    "messages": [{"role": "user", "content": "Hello"}]
  }'

Tool Calling

Declare tools in the tools array of a Responses request. Web search and image generation are supported today — check the model list to confirm the model you use has the capability enabled.

Web Search

curl https://ai.pcpc.ai/v1/responses \
  -H "Authorization: Bearer $PCPC_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "grok-4.6",
    "input": "Search today's news",
    "tools": [{"type": "web_search"}]
  }'

Image Generation

curl https://ai.pcpc.ai/v1/responses \
  -H "Authorization: Bearer $PCPC_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "grok-4.6",
    "input": "Generate an image of a sunset coastline",
    "tools": [{"type": "image_generation"}]
  }'

Developer Tools

Replace YOUR_API_KEY and YOUR_MODEL_ID with real values.

Codex

Codex connects to the gateway over the OpenAI Responses protocol.

Install Codex CLI

curl -fsSL https://chatgpt.com/codex/install.sh | sh

Configure ~/.codex/config.toml

model = "YOUR_MODEL_ID"
model_provider = "pcpc"

[model_providers.pcpc]
name = "PcPc.AI"
base_url = "https://ai.pcpc.ai/v1"
env_key = "PCPC_API_KEY"
wire_api = "responses"

Set your key and start

export PCPC_API_KEY="YOUR_API_KEY"
codex

Claude Code

Claude Code connects to the gateway over the Anthropic Messages protocol. Replace PCPC_API_KEY in the config with your actual API key.

Install Claude Code

curl -fsSL https://claude.ai/install.sh | bash

Configure ~/.claude/settings.json, replacing PCPC_API_KEY below

{
  "env": {
    "ANTHROPIC_AUTH_TOKEN": "PCPC_API_KEY",
    "ANTHROPIC_BASE_URL": "https://ai.pcpc.ai/anthropic",
    "API_TIMEOUT_MS": "6000000",
    "ANTHROPIC_DEFAULT_SONNET_MODEL": "deepseek-v4-flash[1m]",
    "ANTHROPIC_DEFAULT_OPUS_MODEL": "deepseek-v4-flash[1m]",
    "ANTHROPIC_DEFAULT_HAIKU_MODEL": "deepseek-v4-flash[1m]"
  },
  "model": "opus"
}

Start Claude Code

claude

OpenClaw

OpenClaw connects to the Responses endpoint via a custom model provider.

Install OpenClaw

curl -fsSL https://openclaw.ai/install.sh | bash

Configure ~/.openclaw/openclaw.json

{
  agents: {
    defaults: {
      model: { primary: "pcpc/YOUR_MODEL_ID" }
    }
  },
  models: {
    mode: "merge",
    providers: {
      pcpc: {
        baseUrl: "https://ai.pcpc.ai/v1",
        apiKey: "${PCPC_API_KEY}",
        api: "openai-responses",
        models: [
          { id: "YOUR_MODEL_ID", name: "YOUR_MODEL_ID" }
        ]
      }
    }
  }
}

Set your key and start

export PCPC_API_KEY="YOUR_API_KEY"
openclaw chat

Errors & Debugging

Status	Cause	What to do
400	Malformed request or unsupported tool	Check the request body and the model's tool support
401	Invalid API key	Re-copy or create a new API key
402	Plan expired	Purchase a new plan
429	Weekly quota exhausted	Wait for the quota reset or upgrade your plan
5xx	Gateway or upstream error	Retry later and keep the request logs

Never expose your API key in browser front-ends or public repositories. When debugging, log the status code, response body and timestamp — never the full key.

Questions? Email us at admin@pcpc.ai or join the Telegram group.