Quick Start Guide

from openai import OpenAI

client = OpenAI(
    api_key="cf-sk-your-key-here",
    base_url="https://api.clawfeeder.ai/v1"
)

response = client.chat.completions.create(
    model="claude-sonnet-4-6",
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "cf-sk-your-key-here",
  baseURL: "https://api.clawfeeder.ai/v1",
});

const response = await client.chat.completions.create({
  model: "claude-sonnet-4-6",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

curl https://api.clawfeeder.ai/v1/chat/completions \
  -H "Authorization: Bearer cf-sk-your-key-here" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "DeepSeek-V3.2",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from openai import OpenAI
import base64

client = OpenAI(
    base_url="https://api.clawfeeder.ai/v1",
    api_key="cf-sk-your-key-here",
)

# Text-to-image generation
resp = client.images.generate(
    model="gpt-image-2",
    prompt="A red apple on a white table, studio lighting",
    size="1024x1024",
    quality="low",   # low / medium / high — affects credits
    n=1,
)

# Save the result (clawfeeder returns base64 by default)
img_bytes = base64.b64decode(resp.data[0].b64_json)
with open("apple.png", "wb") as f:
    f.write(img_bytes)

curl https://api.clawfeeder.ai/v1/images/generations \
  -H "Authorization: Bearer cf-sk-your-key-here" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "A red apple on a white table",
    "size": "1024x1024",
    "quality": "low",
    "n": 1
  }'

curl https://api.clawfeeder.ai/v1/images/edits \
  -H "Authorization: Bearer cf-sk-your-key-here" \
  -F "model=gpt-image-2" \
  -F "image=@./input.png" \
  -F "prompt=Make the apple green" \
  -F "size=1024x1024" \
  -F "n=1"

# Nano Banana (Gemini native image generation)
curl "https://api.clawfeeder.ai/v1beta/models/nano-banana-2:generateContent" \
  -H "Authorization: Bearer cf-sk-your-key-here" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [
      { "parts": [ { "text": "A red apple on a white table, studio lighting" } ] }
    ],
    "generationConfig": {
      "imageConfig": { "imageSize": "2048x2048" }
    }
  }'

from google import genai
from google.genai import types

# Point the Gemini SDK at clawfeeder
client = genai.Client(
    api_key="cf-sk-your-key-here",
    http_options=types.HttpOptions(base_url="https://api.clawfeeder.ai"),
)

resp = client.models.generate_content(
    model="nano-banana-pro",  # or nano-banana-2
    contents="A red apple on a white table, studio lighting",
    config=types.GenerateContentConfig(
        image_config=types.ImageConfig(image_size="2048x2048"),  # 1K / 2K / 4K → credits
    ),
)
# Image parts are returned inline_data (base64) in resp.candidates[0].content.parts

# 1) Submit text-to-video — returns a task_id immediately
curl https://api.clawfeeder.ai/v1/video/generations \
  -H "Authorization: Bearer cf-sk-your-key-here" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "wan2.6-t2v",
    "input": {
      "prompt": "A small white cat walking on a wooden floor"
    },
    "parameters": {
      "resolution": "720P",
      "duration": 5
    }
  }'
# → {"output":{"task_id":"<id>","task_status":"PENDING"}}

# wan2.6 image-to-video: same endpoint, model=wan2.6-i2v, add input.img_url
curl https://api.clawfeeder.ai/v1/video/generations \
  -H "Authorization: Bearer cf-sk-your-key-here" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "wan2.6-i2v",
    "input": {
      "prompt": "the cat begins to run",
      "img_url": "https://example.com/cat.png"
    },
    "parameters": {"resolution": "720P", "duration": 5}
  }'

# HappyHorse 1.0 i2v / r2v: use input.media: [{"url": "..."}] (array form)
curl https://api.clawfeeder.ai/v1/video/generations \
  -H "Authorization: Bearer cf-sk-your-key-here" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "happyhorse-1.0-i2v",
    "input": {
      "prompt": "camera slowly zooms in",
      "media": [{"url": "https://example.com/photo.jpg"}]
    },
    "parameters": {"resolution": "720P", "duration": 5}
  }'

# HappyHorse 1.0 r2v (reference-to-video, subject/style transfer)
curl https://api.clawfeeder.ai/v1/video/generations \
  -H "Authorization: Bearer cf-sk-your-key-here" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "happyhorse-1.0-r2v",
    "input": {
      "prompt": "the subject dances gracefully",
      "media": [{"url": "https://example.com/reference.jpg"}]
    },
    "parameters": {"resolution": "720P", "duration": 5}
  }'

# 2) Poll status — returns SUCCEEDED (with video URL) or FAILED
curl https://api.clawfeeder.ai/v1/video/generations/<task_id> \
  -H "Authorization: Bearer cf-sk-your-key-here"
# PENDING / RUNNING → keep polling every ~15s
# SUCCEEDED → output.video_url / output.video_duration
# FAILED   → output.message + full refund (you are NOT charged)

import time, requests

BASE = "https://api.clawfeeder.ai"
KEY  = "cf-sk-your-key-here"

# Submit (Aliyun-style nested body — input/parameters)
r = requests.post(
    f"{BASE}/v1/video/generations",
    headers={"Authorization": f"Bearer {KEY}", "Content-Type": "application/json"},
    json={
        "model": "wan2.6-t2v",
        "input": {"prompt": "A small white cat walking on a wooden floor"},
        "parameters": {"resolution": "720P", "duration": 5},
    },
)
task_id = r.json()["output"]["task_id"]
print("task:", task_id)

# Poll until terminal — generation takes ~1-3 min for 5s clips
while True:
    q = requests.get(
        f"{BASE}/v1/video/generations/{task_id}",
        headers={"Authorization": f"Bearer {KEY}"},
    ).json()
    status = q["output"]["task_status"]
    print("status:", status)
    if status in ("SUCCEEDED", "FAILED"):
        print(q)
        break
    time.sleep(15)

curl https://api.clawfeeder.ai/v1/chat/completions \
  -H "Authorization: Bearer cf-sk-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "auto",
    "messages": [
      {"role": "user", "content": "Write a Python fibonacci function"}
    ]
  }'

# Response headers expose the routing decision:
# X-Clawfeeder-Auto-Category: code
# X-Clawfeeder-Auto-Quality:  best
# X-Clawfeeder-Model:         claude-opus-4-7

{
  "model": "auto",
  "quality": "fast",
  "messages": [{"role": "user", "content": "..."}]
}