OpenRouter alternatives (2026)—cheaper, multimodal, with Pix, your own keys, or open source

People looking for an OpenRouter alternative usually have one of four reasons: pay less per token (OpenRouter passes through provider list prices and charges a 5.5 % fee, minimum US$ 0.80, each time you buy credits by card), use image, video, and music models with the same key, put your own provider keys behind a control layer you operate, or pay with Pix, which OpenRouter does not mention. This comparison matches each reason with the tool that addresses it, including cases where staying with OpenRouter is the best choice.

The short answer: to pay less for Claude, Kunavo offers all Claude models 30% below Anthropic’s list price (Claude Sonnet 5 at $1.40 / $7.00 per million input / output tokens, compared with $2.00 / $10.00) and accepts Pix. For self-hosted open source, LiteLLM; for routing and observability over your own keys, Portkey or Helicone; for open-weight models at scale, Together AI.

We disclose the bias up front: Kunavo is our product, which is why it appears first. Every claim about Kunavo is backed by figures on the Kunavo vs OpenRouter page, and every tool on the list gets a genuine recommendation.

The shortlist

One distinction clears up almost everything: inference gateways (OpenRouter, Kunavo, AIMLAPI) resell model access with a single balance, while bring-your-own-key tools (LiteLLM, Portkey, Helicone, Cloudflare) add control over provider accounts you already have. You can combine them, and many teams use one of each. The pattern is explained in the LLM gateway guide (in English).

What the same work costs

Cost of 10 million input tokens and 1 million output tokens with Claude, without prompt caching. The OpenRouter column is the pass-through list price plus the 5.5 % fee for buying credits by card.

With cache hits or batch processing, both columns change. Kunavo’s rates are calculated from the catalog at each build; OpenRouter’s are the list prices from its model list, checked on September 30, 2026.

1. Kunavo — Claude below list price, multimodal with one key, Pix

Kunavo overlaps with OpenRouter in chat—Claude and GPT behind an OpenAI-compatible endpoint—and differs in three respects. Price: Claude is 30% below Anthropic’s list price, and Claude Sonnet 5 costs $1.40 / $7.00 (list price $2.00 / $10.00), with no top-up fees. Modalities: image (Nano Banana, GPT-Image-2), video (Veo 3), and music (Suno, which is not in OpenRouter’s model list) are dedicated endpoints using the same key. Payment: Stripe Checkout with Pix in Brazil, cards, Apple Pay, Google Pay, and Link; prepaid starting at $10, no subscription, and the balance does not expire.

The trade-off is catalog breadth: Kunavo selects leading models it can serve below list price instead of offering hundreds of niche models. It has no free models, runs on shared capacity without a contractual SLA, and operates in a single region in the eastern United States. Run the numbers with the LLM cost calculator or see the point-by-point comparison at Kunavo vs OpenRouter.

2. LiteLLM — self-hosted open source

LiteLLM is a Python proxy that you run yourself: an OpenAI-compatible endpoint in front of your own provider keys, with model routing, per-key budgets, and logs. It is the default answer to “open-source alternative to OpenRouter,” with one important caveat: it does not resell anything. You contract with providers, pay them directly, and operate the proxy. It fits when compliance or procurement requires your own contracts. Details: Kunavo vs LiteLLM.

3. Portkey — the hosted control layer over your keys

Portkey is a hosted gateway over your own provider keys: conditional routing, fallbacks, retries, guardrails, and request analytics, without operating infrastructure. Billing with providers remains direct; Portkey charges for the platform. It suits teams that want to align many internal applications under a single policy layer. Details: Kunavo vs Portkey.

4. Helicone — observability first

Helicone approaches the same space through logs: as a proxy or via an asynchronous SDK, it adds per-request traces, cost attribution, caching, and evaluations to your existing provider traffic. If the problem is “we don’t know where our LLM spend is going,” start here. Details: Kunavo vs Helicone.

5. Together AI — open weights at scale

Together hosts open-weight models — Llama family chat, embeddings, and image — on its own serverless and dedicated infrastructure. It is less a router than a place to run open models quickly, and the right choice if your strategy is open weights rather than a frontier API. Details: Kunavo vs Together AI.

6. AIMLAPI — media-focused aggregator

AIMLAPI aggregates a broad catalog of text and media models behind one API and matches OpenRouter in breadth and image and video coverage. Compare its pricing model and payment terms with your usage pattern. Details: Kunavo vs AIMLAPI.

7. Cloudflare AI Gateway — the edge proxy

Cloudflare AI Gateway sits at the edge in front of the provider keys you already have: response caching, rate limiting, retries, and analytics. It does not resell inference; it is a pure control layer that works with any inference source, including OpenRouter and Kunavo. Details: Kunavo vs Cloudflare AI Gateway.

Paying from Brazil: Pix, cards, and IOF

OpenRouter’s FAQ lists major credit cards, AliPay, and cryptocurrency in USDC (as of September 30, 2026; PayPal is “coming soon”). Pix is not listed. Buying credits by card costs 5.5 % (minimum US$ 0.80), and by USDC costs 5 %; for international cards, your bank’s exchange rate and IOF tax also apply.

At Kunavo, Stripe Checkout displays Pix for customers paying from Brazil: the dollar amount is converted to reais on the page itself, and you scan a Pix QR code without needing a card enabled for international purchases. Cards, Apple Pay, Google Pay, and Link are also available, with no Kunavo top-up fee; Stripe’s exchange rate includes a 2% to 4% conversion fee paid by the buyer, which does not apply when paying in dollars. The step-by-step guide is paying for the API with Pix, and the prices for all Claude models are in Claude API pricing.

When OpenRouter is still the right choice

• You need the broadest catalog. Hundreds of models from more than 80 providers (OpenRouter’s pricing page, as of September 30, 2026): none of the alternatives listed here fully reproduces this breadth.
• You want free models. The :free variants and the openrouter/free router are limited to 50 requests per day, or 1,000 after purchasing at least US$10 in credits (according to the FAQ). Kunavo has none.
• Your work can wait. The OpenRouter model list includes :batch variants of Claude models at half the list price (checked September 30, 2026). For batch processing without urgency, it is cheaper than any discount on real-time calls; Kunavo has no batch endpoint.
• You bring your own provider keys. With pay-as-you-go, the first US$ 25,000 of list-price inference per month is fee-free; above that, 5 % (same FAQ).
• You pay in USDC. OpenRouter accepts USDC with a 5 % fee; Kunavo accepts Pix, cards, and other local methods, while crypto (stablecoins) appears only in the US Checkout, in dollars.

Switching takes two lines

Every serious alternative speaks the OpenAI protocol, so migrating means changing base_url and the key. Only the model ID changes: OpenRouter's provider-prefixed ID (anthropic/claude-sonnet-5) becomes the unprefixed ID (claude-sonnet-5).

import os
from openai import OpenAI

# OpenRouter, Kunavo e LiteLLM falam o mesmo protocolo da OpenAI
client = OpenAI(
    api_key=os.environ["KUNAVO_API_KEY"],   # antes: OPENROUTER_API_KEY
    base_url="https://api.kunavo.com/v1",    # antes: https://openrouter.ai/api/v1
)

resp = client.chat.completions.create(
    model="claude-sonnet-5",   # antes: anthropic/claude-sonnet-5
    messages=[{"role": "user", "content": "Hello"}],
)
print(resp.choices[0].message.content)

What a gateway should solve for you—routing, failover, caching, spending limits—is covered in the OpenAI-compatible API guide (in English).

Frequently asked questions