OpenAI API: every GPT model on one key
Every GPT model OpenAI ships to its API, 29 on Revana today, through one OpenAI-compatible endpoint. Point any OpenAI SDK at https://api.revana.si/v1, name the model, and pay per token from prepaid credit, up to 63% below OpenAI’s list price on the newest models. A request that gets no answer costs $0.00.
GPT-6.1 Sol $0.736 in / $3.68 out per 1M tokens, 63% below OpenAI’s $2.00 / $10.00
GPT-5.5 $1.84 in / $11.04 out per 1M tokens, 63% below OpenAI’s $5.00 / $30.00
GPT-6.1 Terra $0.368 in / $1.84 out per 1M tokens, 63% below OpenAI’s $1.00 / $5.00
Revana prices as of November 2026; OpenAI list prices read 2026-11-16 from developers.openai.com.
Every GPT model and what it costs
USD per million tokens on Revana as of November 2026. Click a model for its full price sheet, cache-write rates and code. Reasoning and thinking tokens bill as output.
GPT-6
The newest generation, each with a 1M-token context window or close to it. 5 models on Revana, USD per million tokens.
| Model | Input | Cached input | Output | Context | OpenAI list (in / out) | Revana vs list |
|---|---|---|---|---|---|---|
GPT-6.1 Terragpt-6.1-terra | $0.368 | $0.0184 | $1.84 | 1.05M | $1.00 / $5.00 | 63% below |
GPT-6.1 Solgpt-6.1-sol | $0.736 | $0.0368 | $3.68 | 1.05M | $2.00 / $10.00 | 63% below |
GPT-6 Solgpt-6-sol | $0.736 | $0.0736 | $3.68 | 1.05M | $2.00 / $10.00 | 63% below |
GPT-6 Lunagpt-6-luna | $0.144 | $0.0144 | $0.72 | 922K | – | – |
GPT-6 Astragpt-6-astra | $3.68 | $0.368 | $18.40 | 1.05M | $10.00 / $50.00 | 63% below |
GPT-5
GPT-5 and its point releases, including mini, nano and Codex models. 13 models on Revana, USD per million tokens.
| Model | Input | Cached input | Output | Context | OpenAI list (in / out) | Revana vs list |
|---|---|---|---|---|---|---|
GPT-5.6 Terragpt-5.6-terra | $0.736 | $0.0736 | $4.416 | 1.05M | $2.00 / $12.00 | 63% below |
GPT-5.6 Solgpt-5.6-sol | $1.472 | $0.1472 | $7.36 | 1.05M | $4.00 / $20.00 | 63% below |
GPT-5.6 Lunagpt-5.6-luna | $0.288 | $0.0288 | $1.728 | 922K | – | – |
GPT-5.5gpt-5.5 | $1.84 | $0.184 | $11.04 | 1.05M | $5.00 / $30.00 | 63% below |
GPT-5.4 nanogpt-5.4-nano | $0.288 | $0.0288 | $1.80 | 128K | – | – |
GPT-5.4 minigpt-5.4-mini | $1.08 | $0.108 | $6.48 | 128K | – | – |
GPT-5.4gpt-5.4 | $3.60 | $0.36 | $21.60 | 1M | – | – |
GPT-5.3 Codexgpt-5.3-codex | $2.52 | $2.52 | $20.16 | – | – | – |
GPT-5.2gpt-5.2 | $2.52 | $0.252 | $20.16 | 128K | – | – |
GPT-5.1gpt-5.1 | $1.80 | $0.18 | $14.40 | 128K | – | – |
GPT-5 nanogpt-5-nano | $0.072 | $0.0072 | $0.576 | 128K | – | – |
GPT-5 minigpt-5-mini | $0.36 | $0.036 | $2.88 | 128K | – | – |
GPT-5gpt-5 | $1.80 | $0.18 | $14.40 | 128K | – | – |
GPT-4.1 and GPT-4o
The previous generation, kept for prompts already tuned to it. 5 models on Revana, USD per million tokens.
| Model | Input | Cached input | Output | Context | OpenAI list (in / out) | Revana vs list |
|---|---|---|---|---|---|---|
GPT-4o minigpt-4o-mini | $0.216 | $0.216 | $0.864 | – | – | – |
GPT-4ogpt-4o | $3.60 | $1.80 | $14.40 | 128K | – | – |
GPT-4.1 nanogpt-4.1-nano | $0.144 | $0.036 | $0.576 | 128K | – | – |
GPT-4.1 minigpt-4.1-mini | $0.576 | $0.144 | $2.304 | 128K | – | – |
GPT-4.1gpt-4.1 | $2.88 | $0.72 | $11.52 | 128K | – | – |
Reasoning models
The o-series, which reason before they answer; reasoning tokens bill as output. 6 models on Revana, USD per million tokens.
| Model | Input | Cached input | Output | Context | OpenAI list (in / out) | Revana vs list |
|---|---|---|---|---|---|---|
o4-minio4-mini | $1.584 | $0.396 | $6.336 | 128K | – | – |
o3-mini (medium)o3-mini-medium | $1.584 | $1.584 | $6.336 | – | – | – |
o3-mini (low)o3-mini-low | $1.584 | $1.584 | $6.336 | – | – | – |
o3-mini (high)o3-mini-high | $1.584 | $1.584 | $6.336 | – | – | – |
o3-minio3-mini | $1.584 | $0.792 | $6.336 | 128K | – | – |
o3o3 | $2.88 | $2.88 | $11.52 | – | – | – |
Revana against OpenAI’s list price
The 7 GPT models where both of Revana’s per-token prices are below OpenAI’s own list. The last two columns price 1,000 requests that each read 3,000 tokens and write 500, a typical chat or support call. OpenAI list prices from https://developers.openai.com/api/docs/pricing, read 2026-11-16.
| Model | OpenAI list (in / out) | Revana (in / out) | Below by | 1,000 calls at OpenAI | 1,000 calls on Revana |
|---|---|---|---|---|---|
| GPT-6.1 Terra | $1.00 / $5.00 | $0.368 / $1.84 | 63% | $5.50 | $2.02 |
| GPT-6.1 Sol | $2.00 / $10.00 | $0.736 / $3.68 | 63% | $11.00 | $4.05 |
| GPT-5.5 | $5.00 / $30.00 | $1.84 / $11.04 | 63% | $30.00 | $11.04 |
| GPT-6 Sol | $2.00 / $10.00 | $0.736 / $3.68 | 63% | $11.00 | $4.05 |
| GPT-6 Astra | $10.00 / $50.00 | $3.68 / $18.40 | 63% | $55.00 | $20.24 |
| GPT-5.6 Terra | $2.00 / $12.00 | $0.736 / $4.416 | 63% | $12.00 | $4.42 |
| GPT-5.6 Sol | $4.00 / $20.00 | $1.472 / $7.36 | 63% | $22.00 | $8.10 |
Older GPT models are not in this table because they cost more on Revana than at OpenAI. Same model, same answer: Revana passes the request to OpenAI’s model unchanged.
Match the GPT model to the job
What a workload costs on Revana at today’s prices, worked out from the token rates above.
High-volume classification
100,000 short labels on texts of about 600 tokens: GPT-5 nano: $5.47 · GPT-6.1 Terra: $25.76. GPT-5 nano costs 79% less for this job. Return one label with a low max_tokens, and check a sample against a bigger model before the full run.
Code work
100 code edits that each read 30K tokens (20K cached) and write 3K: GPT-5.3 Codex: $13.61 · GPT-6.1 Sol: $1.91. GPT-6.1 Sol costs 86% less for this job. Keep the repository context first in the prompt so later calls read it from the cache.
Long context
1 question about a 400K-token document: GPT-5.5: $1.51 · GPT-6.1 Sol: $0.60. GPT-6.1 Sol costs 60% less for this job. Past 272K prompt tokens the GPT models that have a long-context tier bill at their higher rate; the cost here includes it.
Step-by-step reasoning
100 reasoning tasks that read 5K tokens and think and write 8K: o4-mini: $5.86 · GPT-6 Astra: $16.56. o4-mini costs 65% less for this job. Reasoning tokens bill as output, so set max_tokens to the longest answer you will accept.
Call GPT with the code you already have
Two settings change: the base URL and the key. Messages, tools, response_format and streaming stay in the OpenAI format, and switching to another GPT model, or another maker’s, is a change to the model field.
curl https://api.revana.si/v1/chat/completions \
-H "Authorization: Bearer $REVANA_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model":"gpt-6.1-sol","messages":[{"role":"user","content":"Write a haiku about routing."}]}'import os
from openai import OpenAI
client = OpenAI(base_url="https://api.revana.si/v1", api_key=os.environ["REVANA_API_KEY"])
reply = client.chat.completions.create(
model="gpt-6.1-sol",
messages=[{"role": "user", "content": "Write a haiku about routing."}],
)
print(reply.choices[0].message.content)
print(reply.usage.cost_usd) # what this request cost, in USDimport OpenAI from "openai"
const client = new OpenAI({ baseURL: "https://api.revana.si/v1", apiKey: process.env.REVANA_API_KEY })
const reply = await client.chat.completions.create({
model: "gpt-6.1-sol",
messages: [{ role: "user", content: "Write a haiku about routing." }],
})
console.log(reply.choices[0].message.content, reply.usage.cost_usd)The answer is OpenAI’s chat.completion object, and usage.cost_usd is the exact amount taken from your balance. For 3,000 tokens in and 500 out on GPT-6.1 Sol:
{
"id": "chat_3f9c…",
"object": "chat.completion",
"model": "gpt-6.1-sol",
"choices": [
{
"index": 0,
"message": {
"role": "assistant",
"content": "…"
},
"finish_reason": "stop"
}
],
"usage": {
"prompt_tokens": 3000,
"completion_tokens": 500,
"total_tokens": 3500,
"prompt_tokens_details": {
"cached_tokens": 0
},
"cost_usd": 0.004048
}
}Send "stream": true for server-sent events, and "stream_options": {"include_usage": true} to get the cost in the last chunk. Every field is in the chat completions docs.
Pay only for the answers you get
- Automatic fallback. A model with more than one route starts on its lowest-cost route; a busy or failed route hands the request to the next before any answer is sent. Each model page shows its routes.
- Held, then charged. While a request runs, Revana holds its prompt plus
max_tokens(8,192 output tokens when you set none). When it ends the hold is released and only the tokens used are charged. - No answer costs $0.00. If every route fails you get an OpenAI-shaped error and no charge;
503 model_busyis never billed, so a short retry is safe. - Backup routes stay under the hold. When a model’s listed route is busy, a backup can cost more; you are charged that route’s cost, never more than the hold, and it is always in
cost_usd.
Make images and video from Claude Code, Cursor and ChatGPT
- Connect the Revana MCP server to Claude Code, Claude Desktop, claude.ai, Cursor, VS Code, Codex CLI or ChatGPT, and your agent can quote, generate and download images and video on your credit.
- Every language model and its price is published as llms.json and through
GET /v1/models, so code can pick a model by price at run time. - Any tool that sends HTTP works too, such as n8n’s HTTP Request node:
POST https://api.revana.si/v1/chat/completionswith your key as a Bearer token.
GPT Image on the same balance
OpenAI's image models run on the same Revana key through POST /v1/generations, billed per image from the same credit as your GPT tokens.
| Model | Kind | From |
|---|---|---|
| GPT Image 1.5 | image | $0.01 an image |
| GPT Image 1 | image | $0.01 an image |
| GPT Image 2.5 Flare | image | $0.02 an image |
| GPT Image 2.5 Sunburst | image | $0.02 an image |
Every image, video and audio model and its price is at revana.si/models. A generation that fails costs nothing.
From signup to your first GPT call
- Create a key. Sign up and create an API key in the dashboard. It works with every language, image and video model.
- Add prepaid credit. Packs start at $10 and 100 credits are $1. Credit never expires; no subscription or monthly minimum.
- Send the sample request. Copy the curl above, set your key, and read the answer and its
cost_usd.
FAQ
How do I get an Open AI API key for Revana?
Sign up at revana.si, create an API key in the dashboard and add prepaid credit. Use that key with any OpenAI SDK and the base URL https://api.revana.si/v1; you do not need an OpenAI account.
How much do OpenAI models cost per million tokens?
On Revana, GPT models run from $0.072 in and $0.576 out per million tokens (GPT-5 nano) to $3.60 in and $21.60 out (GPT-5.4), prices as of November 2026.
Is GPT cheaper on Revana than on OpenAI?
For the newer models, yes: GPT-5.5 costs $1.84 in and $11.04 out against $5.00 and $30.00 (63% below); GPT-6.1 Sol costs $0.736 in and $3.68 out against $2.00 and $10.00 (63% below); GPT-6.1 Terra costs $0.368 in and $1.84 out against $1.00 and $5.00 (63% below). OpenAI's list prices were read on 2026-11-16. Older models such as GPT-4.1, GPT-4.1 mini and GPT-4.1 nano cost more on Revana, so keep those on OpenAI's API if price is what matters.
Is it the same model as calling OpenAI directly?
Yes. Revana passes your request to OpenAI's model unchanged and returns OpenAI's chat.completion object, with the usage the model reported plus cost_usd.
Does prompt caching work?
Yes. Automatic prompt caching works as it does at OpenAI, and cached input is billed at each model's cached rate, shown in the table.
Which OpenAI features are not supported?
The chat completions endpoint takes text, pictures as image_url and function tools. The Responses and Assistants APIs, web search, audio, files and built-in tools are not available; those parameters are refused with 400 unsupported_parameter.
Is there a subscription or minimum spend?
No. You pay per token from prepaid credit, packs start at $10, and credit never expires. A request that gets no answer costs nothing.
Start with one API key
One key for every language, image, video and audio model on Revana. Prepaid credit, no subscription, and a request that fails costs nothing.