Multi-Model AI Gateway · Pay-as-you-go · Global Access
Access Kimi K3 · Kimi K2.7 Code · Kimi K2.7 HighSpeed · GLM-4 Flash
Unified API · Pay-as-you-go · Millisecond Response · Global Access
✅ 4 models available now · Instant access · No credit card · 1M tokens free for 14 days
Type a message and see the response. No account needed.
Live demo, no account needed — 5 messages per hour, single-turn. Your own free key: 1,000,000 tokens, 14 days, no card.
Updated live · Refreshes every 30 min
Not ready to pay yet? Start with 1M free tokens →
Connect Claude, Cursor, Windsurf and any MCP client to 6 Chinese AI models in one line
See how much you save vs OpenRouter
Based on Kimi K3 list pricing: $3.50 per 1M fresh input tokens, $18.00 per 1M output tokens. The OpenRouter figure applies the ~50% markup it adds on Chinese models; TokenEase bills upstream cost.
We don't mark up model costs. Here is exactly what we pay upstream — and how we pass the savings on to you.
| Model | Input (cached) | Input (fresh) | Output | Context | Source |
|---|---|---|---|---|---|
| Kimi K3 | $0.28 / 1M | $3.50 / 1M | $18.00 / 1M | 256K | Moonshot |
| Kimi K2.7 Code | $0.30 / 1M | $1.20 / 1M | $5.00 / 1M | 256K | Moonshot |
| Kimi K2.7 Code HighSpeed | $0.50 / 1M | $2.50 / 1M | $10.00 / 1M | 256K | Moonshot |
| GLM-4 Flash | — | $0.10 / 1M | $0.40 / 1M | 128K | Zhipu |
Our catalogue holds 17 models. The other 13 — DeepSeek, Qwen, Doubao, the paid GLM tiers and Tencent Hunyuan — stay listed at
GET /v1/models with enabled: false until their upstream billing is settled.
Always read /v1/models; never hard-code a model name.
💡 How we price: Our plans bundle any enabled model at flat monthly rates ($9.9 / $29.9 / $99), and metered overage is billed at the upstream rates above — no peak surcharge, no markup on tokens.
Early access users from 12+ countries
"Switched from OpenRouter — same models, cheaper. Setup took 5 minutes, the unified API just works."
— Alex K., Full-Stack Developer Early Access
"6 Chinese AI models + Kimi K3 long-context in one key. Way better than juggling 6 vendor accounts."
— Marcus L., AI Engineer Early Access
"The model pricing transparency is unmatched. I know exactly what I'm paying per call. Saved me $200/month."
— Priya S., Indie Hacker Early Access
🔒 Security & Compliance
🔐 TLS 1.3 Encrypted 🛡️ SOC 2 Type II Ready 🌐 GDPR Compliant 💳 Paddle PCI DSS 🚫 Zero Data Retention
Trusted by developers in 12+ countries · 99.9% uptime SLA · 24h support response
Get 1M tokens free for 14 days — no credit card required
Have questions? Reach out through the following channels:
curl -X POST https://tokenease.io/v1/chat/completions \
-H "Authorization: Bearer YOUR_TOKEN_HERE" \
-H "Content-Type: application/json" \
-d '{
"model": "kimi-k3",
"messages": [{"role": "user", "content": "Hello"}]
}'
💡 Switching from OpenRouter? See detailed price comparison and migration guide.
TokenEase vs OpenRouter →📊 Comparing Chinese LLM gateways? See how TokenEase stacks up against ChinaModelAPI, tokencnn, and OpenRouter.
Full Competitor Comparison →