● Per-token pricing · OpenAI-compatible

Open models at low per-token prices

DeepSeek, Qwen, Llama and more behind one OpenAI-compatible API. Pay only for the tokens you use.

BillingPer token
APIOpenAI-compatible
Models8

Featured models

USD per 1M tokens.

ModelCreatorTypeContextInput / 1MOutput / 1M
Qwen3 32BQwenchat40K$0.10$0.30
Llama 3.3 70BMetachat128K$0.23$0.40
DeepSeek V3DeepSeekchat128K$0.27$1.10
DeepSeek R1DeepSeekchat128K$0.55$2.19

View all models →

Switch in two lines

Point any OpenAI SDK at Keyra: change the base URL and the key.

curl

curl https://api.keyra.example/v1/chat/completions \
  -H "Authorization: Bearer $KEYRA_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "qwen3-32b", "messages": [{"role": "user", "content": "Hello"}]}'

Python · openai SDK

import os
from openai import OpenAI

client = OpenAI(base_url="https://api.keyra.example/v1", api_key=os.environ["KEYRA_API_KEY"])

reply = client.chat.completions.create(
    model="qwen3-32b",
    messages=[{"role": "user", "content": "Hello"}],
)
print(reply.choices[0].message.content)

Pay per token

No minimums, no seats. Input and output priced separately.

OpenAI-compatible

Keep your SDK and your code. Change the base URL.

Published prices

Every price is on this site, with the date it last changed.

Start building