Claude, GPT and media models,Up to 70% off official rates.
Kunavo is an independent, OpenAI-compatible AI API gateway: one API key and one pay-as-you-go balance reach Claude, GPT and image, video and audio models, with no per-provider account required.
Use Kunavo as your model provider — an OpenAI-compatible gateway to every frontier text, image and video model. base_url: https://api.kunavo.com/v1 auth: Authorization: Bearer $KUNAVO_API_KEY To use a model, call GET /v1/models for the live catalog, then route each model by its kunavo.endpoint field. Full agent reference: https://kunavo.com/llms.txt
Three ways to start
Buy API access
Claude and GPT through one key and one prepaid balance — from $10, no subscription.
USD per 1M tokens, input / output · pick one to sign up with it
Move from another provider
Code on an OpenAI SDK keeps working: change base_url and the key. From OpenRouter, model ids drop the vendor/ prefix — the comparison lists what carries over and what doesn't.
# base_url - https://api.openai.com/v1 + https://api.kunavo.com/v1
Connect your app
Setup pages for coding agents, chat apps and workflow tools, each checked against the client's own docs and dated.
Providers we’ve unified
The AI gateway built for builders who ship.
From the routing layer to the billing ledger, every part of Kunavo was designed for indie developers and small teams shipping AI features for real customers.
Edge-fronted, US-East hosted
Anycast edge network terminates TLS close to you; the gateway itself runs in a single US-East region (Ashburn, Virginia).
OpenAI-compatible
Drop-in replacement for OpenAI SDKs. Streaming, function calling, tool use, vision — all wire-compatible. No new client to learn.
Prepaid balance, pay as you go
One prepaid balance in US dollars pays for every model. Self-serve top-ups, opt-in auto-recharge, no subscription.
Frontier models, up to 70% off
Models from OpenAI, Anthropic and Google, most priced below the provider’s official rate — up to 70% off depending on the model. Claude, GPT, Nano Banana, Veo — text, image and video, one balance.
Transparent pricing
Every model’s per-1M-token price is published. No hidden multipliers, no surprise overages. Failed requests are never billed.
Automatic failover
Every request can fall through up to three upstream channels. A channel that has been failing is bypassed before your request is sent — no client-side retry needed.
First-class streaming
Native SSE pass-through. Time-to-first-token matches the upstream provider — no buffering, no batching, no delay.
Granular usage data
Per-call analytics by model, key and IP. Webhook deliveries when an image, video or music task finishes. Export everything as CSV when you need it.
Prompt caching, up to 90% off
Cache reads bill at 10% of input on Claude and GPT, 5% on Claude Opus 5.5 and 2.5% on Claude Fable 5.1 — pass cache_control on your system prompt and long context becomes a near-free re-read. Hit rate and savings are shown live in your dashboard.
What to build with Kunavo.
- Customer Support
AI customer support
The fastest-ROI AI deployment in any B2C SaaS — automate ticket triage, draft 80% of responses, and escalate the rest cleanly. Production code, real cost numbers, and the compliance pitfalls that catch teams off-guard.
Explore use case - Knowledge Base
RAG chatbot API
Most internal knowledge bases are dead documentation — nobody finds anything. A Claude-backed RAG chatbot turns them into a real assistant that cites sources and refuses when it doesn't know. Here's the production pattern.
Explore use case - Trust & Safety
AI content moderation
Modern moderation isn't just regex — it's nuance: sarcasm, dog whistles, brand-context misuse, image+text combinations. LLMs do this far better than rule-based systems, at a price that scales.
Explore use case - Developer Tools
AI code assistant
Cursor, Aider, Cline, Continue.dev — they're all powered by the same handful of frontier LLMs. If you're building a coding tool (or a co-pilot inside your own dev product), here's the architecture and the cost reality.
Explore use case - Data Processing
AI data extraction
The boring, valuable use case. Invoices, receipts, contracts, leads, resumes — anywhere you'd previously have built a parser, an LLM held to a JSON schema does it in 30 lines, more accurately, and you can ship in a day instead of a quarter.
Explore use case
Frontier models, up to 70% off official.
Claude Fable 5.1
Anthropic's most capable model — frontier reasoning and long-horizon agentic work.
Claude Fable 5
Frontier reasoning and long-horizon agents — the model Fable 5.1 succeeds.
Claude Opus 5.5
Anthropic's newest Opus — long-running agentic coding and knowledge work, for less than Opus 5.
Claude Opus 5
Near-flagship Opus reasoning at half the price of Fable 5 — vision and agentic coding.
Claude Opus 4.8
Anthropic Opus 4.8 — stronger agentic coding and honesty.
Claude Opus 4.7
Anthropic Opus 4.7 — previous-generation Opus; reasoning and vision.
Claude Opus 4.6
Anthropic Opus 4.6 — deep reasoning, exceptional agentic ability.
Claude Sonnet 5
Near-Opus coding and agentic quality at Sonnet cost.
Claude Sonnet 4.6
Balanced speed/quality — the everyday production workhorse, elite coding.
Claude Haiku 4.5
Anthropic Haiku 4.5 — fast and cost-efficient.
GPT-6 Astra
OpenAI's newest flagship — frontier reasoning and long-horizon agentic coding.
GPT-6 Sol
GPT-6 for complex coding and agentic workflows — at a fifth of GPT-6 Astra's price.
GPT-6 Luna
OpenAI's most efficient GPT-6 — focused, high-volume tasks.
GPT-5.6 Sol
OpenAI's newest flagship — top-tier reasoning and agentic coding.
GPT-5.6 Terra
GPT-5.6 mid tier — the everyday workhorse of the 5.6 family.
GPT-5.5
OpenAI GPT-5.5 — flagship reasoning and agentic tool use.
Point your agent at llms.txt
It uses every model itself.
Hand one instruction to Claude Code, Cursor, Cline — or any OpenAI-compatible agent. It reads the live model catalog from Kunavo and drives text, image and video models on its own. No SDK, no glue code.
- OpenAI-wire compatible — agents need no custom integration
- GET /v1/models is the live catalog — never hardcode model names
- One key for every modality: text, image, video, audio
Use Kunavo as your model provider — an OpenAI-compatible gateway to every frontier text, image and video model. base_url: https://api.kunavo.com/v1 auth: Authorization: Bearer $KUNAVO_API_KEY To use a model, call GET /v1/models for the live catalog, then route each model by its kunavo.endpoint field. Full agent reference: https://kunavo.com/llms.txt
The more you pre-pay, the more you save.
Pre-paid wallet. $10 starts you up. No subscription, no minimum, balance never expires.
Starter
Just exploring
- Access to every model and endpoint
- Per-call usage analytics
- Email support
- No subscription, no expiry — top up from $10
Builder
Limited · +$10Shipping a product
- $100 deposit = $110 credit
- 10% bonus, limited time
- Priority email support
- Everything every account gets
Scale
Limited · +$200Running production traffic
- $1000 deposit = $1200 credit
- 20% bonus, limited time
- Priority email support
- Everything every account gets
Enterprise
Limited · +$1250High-volume scale
- $5000 deposit = $6250 credit
- 25% bonus, limited time
- Priority email support
- Everything every account gets
Included with every account
- Unlimited API keys
- Per-key monthly spend limits
- Per-key IP allowlists
- Auto-recharge from a saved card — any whole-dollar amount from $10, opt-in
- Webhooks for image, video and music tasks
- Usage API and CSV export
- OpenAI-compatible and Anthropic Messages endpoints
Start with the popular guides.
- Integration
Claude Code API key — where to get one, where to put it, and why the wrong variable fails silently
Claude Code takes a subscription login or an API key, and the two behave very differently once set. Here is where each comes from, exactly which variable holds it, and the header mismatch behind most 401s.
Read - Setup
Claude API key
Get a Claude key and call Claude via the OpenAI SDK or the native Messages API.
Read - Pricing
Anthropic Claude API pricing October 2026 (official rates)
Claude is the default for reasoning, coding and agents. Here are the current Claude API prices per model, verified October 2026 — Anthropic's official list alongside Kunavo's rates at 30% less — with worked examples and the levers that cut a Claude bill the most.
Read - Pricing
Claude Code pricing 2026 — subscription vs API, per-model rates, and what a month costs
Claude Code costs either a flat monthly fee or a per-token bill, and almost every confusing answer about its price comes from mixing the two up. Here are the current numbers for both routes, what a real month costs, and the arithmetic for choosing.
Read - Pricing
OpenAI GPT API pricing 2026 — GPT-6, GPT-5.6 & GPT-5.5 costs, examples, cheaper access
GPT-5 and GPT-6 are OpenAI's current families. Here are the current GPT API prices per model — 60–65% below OpenAI's list — with worked cost examples, the official OpenAI reference, and the cheapest way to call GPT in production.
Read - Architecture
LLM gateway — one API for every model: routing, fallback, billing & observability (2026)
An LLM gateway is one API in front of every model provider — routing, failover, caching, one bill and per-key spend control. This is what it does, how it differs from an API gateway, the two architectures, and how Kunavo implements it OpenAI-compatibly.
Read - API
OpenAI-compatible API guide — from Ollama's local endpoint to hosted frontier models
'OpenAI-compatible' means any OpenAI SDK works by changing only base_url. Ollama exposes one for local models; Kunavo exposes one for hosted frontier models. Here's the pattern, and how to move between them without rewriting code.
Read - Tools
LLM cost calculator
Estimate per-request and monthly API cost for Claude, Gemini and GPT — with savings vs list.
Read - Tools
Token counter
Paste any text and get its token count, and what it costs as input or output on every Claude and GPT model.
Read - Video
Text-to-video API — generate video from a prompt (OpenAI-compatible, 2026)
A text-to-video API turns a prompt into a video clip programmatically. Here's the API shape, the live models on Kunavo (Veo 3.1, Seedance 2, Wan 2.7), per-clip pricing, and the async pattern you need in production.
Read - Video
Veo 3.1 API — Google's text-to-video with native audio, OpenAI-compatible (2026)
Veo 3.1 is Google's text-to-video model with native audio. Here's how to call the Veo 3.1 API on Kunavo — three tiers, image-to-video, per-video pricing ~50–70% under Google's list, and the async workflow.
Read - Audio
Suno API documentation — generate music from a prompt (V5 / V5.5), with pricing
Suno generates full songs — vocals, lyrics and instrumentation — from a prompt. Here's the Suno API documentation: the /v1/audio/music endpoint, custom lyrics, per-request pricing, and the async workflow.
Read - Image
Nano Banana API — call Google's image models OpenAI-compatibly, 25–50% under list
Nano Banana is Google's image model family. Here's how to call it through an OpenAI-compatible API — text-to-image, image edit, per-image pricing 25–50% under Google's list, on one key.
Read - Image
GPT-Image-2 API — OpenAI's image model at half the price, OpenAI-compatible
GPT-Image-2 is OpenAI's latest image model. Here's how to call the GPT image API through an OpenAI-compatible endpoint — text-to-image, image edit, per-image pricing at about half of list, on one key.
Read - Compare
OpenRouter alternative
Kunavo vs OpenRouter — text breadth vs multimodal, pricing, API keys and payments.
Read - Compare
OpenRouter alternatives (2026) — an honest roundup: cheaper, multimodal, BYO-key and open-source options
OpenRouter routes to hundreds of models — image, video and speech included — and is sometimes still the right call. This is the honest map of alternatives — by motivation: lower prices, specific media models on one key, your own keys under a control plane, or open source you can self-host.
Read - Models
Claude Opus 5.5 API — pricing, model ID and what changed from Opus 5
Anthropic released Claude Opus 5.5 on September 22, 2026 — cheaper than Opus 5 and, by Anthropic's numbers, better. Here is its API price at Anthropic and on Kunavo, the model ID, what the breaking changes do to requests written for Opus 5, and working code for the OpenAI SDK, the Anthropic SDK and Claude Code.
Read - Setup
CC Switch setup
Add Kunavo to CC Switch for Claude Code and Codex — which API format to pick, and why.
Read
Set up Kunavo in the tool you already use
All 53 integrationsRecently published.
- Can Claude generate images? What Claude can and cannot do with images (2026)
- Is Codex free? Yes, with limits — the free tier, the plans, and what a task costs
- Claude card declined? Why Anthropic rejects cards and how to pay anyway
- How to pay for Claude with UPI in India: Claude Pro, the API and Claude Code
- Can you pay for the OpenAI API with UPI? What works from India
- Pay for the Claude and GPT APIs with crypto (USDC)
- Pay for Claude with Cash App: Cash App Pay vs the Cash App Card (2026)
- Pay for Claude with Klarna: what works (and what it costs)
- Pay for Claude with Afterpay: what works and what doesn't (2026)
- Pay for Claude with Affirm: Claude Pro, the API and Claude Code
- Pay for Claude with Zip (Pay in 4): what works and what doesn't
- Pay for Claude with Google Pay: claude.ai, the Android app and the API
- Pay for Claude with Apple Pay: the iPhone app, claude.ai and the API
- Pay for the Claude API with a bank transfer (ACH)
- Pay for Claude by bank in the UK: Pay by Bank and Bacs Direct Debit
- Claude Fable 5.5 — release date, what is confirmed, and the Fable API today
- Kimi Code pricing: membership tiers, Extra Usage and the metered API
- Roo Code vs Continue: both have ended — which setup survives, and where it goes
- Chatbox AI pricing: hosted plans, compute points and when your own key is cheaper
- Mistral Vibe pricing: plan budgets, pay-as-you-go keys and your own provider
- Qoder pricing: what a credit is worth, what each request costs, and when your own key wins
- Warp AI pricing: credits, plans and what your own inference removes
- Best local model for OpenClaw: gemma4, gpt-oss:20b and qwen3-coder:30b measured on four agent tasks
- DeepSeek Harness pricing: the harness is free, the model is not — and what each turn sends
Recent deep dives.
- Practical guide · 5 min
Calling Claude / GPT APIs from China — Connectivity Self-Test, Payments, and Retry Patterns
Call Kunavo from mainland China: the gateway is in a single US East region. Whether you can connect and what latency you experience depend on your network; self-test commands are included. Pay with Alipay, WeChat Pay, or bank cards such as Visa and Mastercard. Robust retry code, plus guidance on when forwarding is actually necessary.
- Tutorial · 6 min
Creating short video ads for Taiwanese brands with Veo 3 (Sora on the same endpoint coming soon) — complete 5-minute tutorial
A complete workflow from text prompts to 9:16 vertical Reels / TikTok / IG short videos. Animate existing product photos for image-to-video generation. Five real Taiwanese brand applications, Traditional Chinese text-rendering considerations, and advanced techniques for commercially usable quality.
- Implementation guide · 8 min
Build a Japanese RAG chatbot with Claude — a 5,000-document knowledge base in 30 lines
A complete RAG implementation that makes 5,000 internal documents searchable with Claude Sonnet 4.6. About 0.9 yen per query after prompt caching. Embeddings are not provided by Kunavo; only that step is billed directly by OpenAI. Includes Japanese-specific token usage, hallucination countermeasures, and a production-readiness checklist.
Everything you’re
wondering about.
Didn’t answer your question? Email us at contact@kunavo.com — we reply within 24 hours.
Kunavo is purpose-built for indie developers and small teams shipping production AI features. Three real differences: (1) we cover text, image, video and music on one balance; (2) one prepaid balance, pay as you go — no subscription, no off-platform invoices; (3) full transparency on routing — we never silently swap your model to a cheaper one.
Most models are priced below the provider’s official list price — up to 70% off depending on the model. Bigger top-ups add a bonus on top. You also save operationally: one account, one balance, one SDK, no commitment minimums. The per-1M-token price for every model is published on /pricing — easy to compare against the upstream listing anytime.
Yes. We implement the full set of OpenAI endpoints: /v1/chat/completions, /v1/embeddings, /v1/images/generations, /v1/models and /v1/video/generations. Streaming, function calling, vision and tool use all behave identically. Projects using the OpenAI SDK migrate by changing base_url — that’s it.
No. Kunavo is a pre-paid wallet. Top-ups stay in your account forever — no subscriptions, no monthly minimums, no expiration. Unused balance you paid for can be refunded to your original payment method at any time, without closing your account, less a proportional processing fee.
Never. 4xx and 5xx responses are not billed. Streaming responses that disconnect mid-flight are billed only for the tokens actually delivered. Every charge is visible per-call in the usage dashboard, exportable as CSV for accounting.
Top-ups go into one prepaid balance in US dollars, and every call is paid from it, with no subscription. The payment page shows the methods available where you are. Auto-recharge is optional and refills the wallet by an amount you choose.
Kunavo runs in one US-East region (Ashburn, Virginia) behind an anycast edge that terminates TLS near you. Accounts, billing and usage data live in a single primary database with daily encrypted volume snapshots.
Three minutes to your first call.
One OpenAI-compatible API for Claude, GPT, Nano Banana, Veo and Suno — $10 minimum top-up, pay only for what you call.