Anthropic doesn't offer a free Claude API directly. On AIgateway, adding a card (no charge) gets you $1 across a curated trial catalog that includes moonshot/kimi-k2.7-code, an open-weight model benchmarked against Claude Opus 4.7 on agentic coding (SWE-Bench 68.2) at ~10x lower cost. Full Claude Opus 4.7 and Sonnet 4.6 are available at pass-through pricing once you top up.
Visit aigateway.sh and sign in with GitHub or Google. No phone verification, no invite codes. A payment method is required to call the API — adding a card is free and never charges you, and grants $1 in credit across a curated trial catalog of 15 models: zai-org/glm-5.2, moonshot/kimi-k2.7-code (frontier open-weight, 256K context, vision, native tool calling), BGE-M3 embeddings, FLUX-2 Klein image, Moondream vision, Aura 2 TTS, Whisper Turbo STT and more. The $1 expires 30 days after you add the card.
Install pip install aigateway-py openai, point the OpenAI SDK at https://api.aigateway.sh/v1, set model="zai-org/glm-5.2", and you're making production-grade LLM calls. Your $1 card credit covers GLM-5.2, model="moonshot/kimi-k2.7-code" and the rest of the curated trial catalog. Standard PAYG rate limits apply.
Top up $5 (or more) for pay-as-you-go. No subscription, no auto-renew. That unlocks the full catalog — Claude Opus 4.7, GPT-5.4, Gemini 3.1 Pro, the image and video models, everything. Your code stays the same.
# pip install aigateway-py openai
# aigateway-py: sub-accounts, evals, replays, jobs, webhook verify.
# openai SDK: chat/embeddings/images/audio — drop-in compat per our SDK's own guidance.
from openai import OpenAI
client = OpenAI(
base_url="https://api.aigateway.sh/v1",
api_key="sk-aig-...",
)
r = client.chat.completions.create(
model="moonshot/kimi-k2.7-code",
messages=[{"role": "user", "content": "Explain vector databases in two sentences."}],
)
print(r.choices[0].message.content)// npm i aigateway-js openai
// aigateway-js: sub-accounts, evals, replays, jobs, webhook verify.
// openai SDK: chat/embeddings/images/audio — drop-in compat.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.aigateway.sh/v1",
apiKey: process.env.AIGATEWAY_KEY,
});
const r = await client.chat.completions.create({
model: "moonshot/kimi-k2.7-code",
messages: [{ role: "user", content: "Explain vector databases in two sentences." }],
});
console.log(r.choices[0].message.content);A payment method is required to call the API — but adding a card is free and never charges you, and it grants $1 in credit across a curated trial catalog of 15 models led by GLM-5.2 (chat, code, image, voice, embeddings, vision). The $1 expires 30 days after you add the card.
No catch. The $1 card credit exists because we believe developers should try the platform end-to-end before paying — the card is a signup gate, not a charge. Most teams upgrade once they've shipped a prototype on the credit; the paid tier starts at a $5 minimum top-up with no subscription.
Yes — Kimi K2.7 Code is a 1T-parameter open-weight model benchmarked at SWE-Bench 68.2 and HumanEval 92.7. It's competitive with Claude Opus 4.7 on agentic coding at ~10x the price-per-token once you're paying.
For prototyping and personal projects, yes. For production traffic, the $1 and the curated catalog are the bottleneck — a $5 top-up unlocks higher throughput across the full catalog at pass-through pricing with no recurring charge.