GPT-5.6 Terra API
Call GPT-5.6 Terra with one TokenMilo API key. The previous GPT generation, still widely used and very low cost. Billed per token at 91% below the official OpenAI price.
| Model ID | gpt-5.6-terra |
|---|---|
| Provider | OpenAI |
| Family | GPT-5.6 |
| Discount | 91% off official |
GPT-5.6 Terra pricing
USD per 1M tokens. Prices updated 2026-11-13.
| TokenMilo | Official OpenAI | You save | |
|---|---|---|---|
| Input | $0.171 | $2 | −91% |
| Output | $1.03 | $12 | −91% |
| Cache read | $0.017 | — | — |
| Cache write (5 min) | $0.216 | — | — |
| Cache write (1 hour) | $0.216 | — | — |
OpenAI cache pricing varies by model, so no official figure is shown for cache rows. 1-hour cache writes are billed at the same price as 5-minute writes.
What it costs in practice
| Workload | TokenMilo | Official | You save |
|---|---|---|---|
| 1,000 requests × 2,000 in / 500 out tokens | $0.855 | $10.00 | $9.14 |
| 1,000 requests × 50,000 in / 2,000 out tokens | $10.60 | $124 | $113 |
GPT-5.6 price comparison: TokenMilo vs official
| Model | TokenMilo in / out | Official in / out | Saved |
|---|---|---|---|
| GPT-5.6 Sol | $0.342 / $2.05 | $4 / $20 | −90% |
| GPT-5.6 Terra (this page) | $0.171 / $1.03 | $2 / $12 | −91% |
| GPT-5.6 Luna | $0.017 / $0.099 | $0.2 / $1.2 | −92% |
GPT-5.6 Terra use cases
Low-cost chat and summarisation
Good-enough quality for everyday chat at a very low price.
Batch jobs and data processing
Cleaning, tagging and transforming large datasets.
Prototyping
Build and test ideas cheaply before moving to a newer model.
Legacy integrations
Projects pinned to GPT-5.x that you don’t want to re-test yet.
Supported endpoints
| OpenAI Chat Completions · recommended | POST https://api.tokenmilo.com/v1/chat/completions |
| OpenAI Responses · recommended | POST https://api.tokenmilo.com/v1/responses |
| Anthropic Messages | POST https://api.tokenmilo.com/v1/messages |
Base URL and API key are the same for every format. API guide →
Code examples
curl
curl https://api.tokenmilo.com/v1/chat/completions \
-H "Authorization: Bearer $TOKENMILO_API_KEY" \
-H "content-type: application/json" \
-d '{
"model": "gpt-5.6-terra",
"messages": [{"role": "user", "content": "Hello, GPT-5.6 Terra!"}]
}'Python
# pip install openai
import os
from openai import OpenAI
client = OpenAI(
base_url="https://api.tokenmilo.com/v1",
api_key=os.environ["TOKENMILO_API_KEY"],
)
res = client.responses.create(
model="gpt-5.6-terra",
input="Explain prompt caching in one paragraph.",
)
print(res.output_text)Use GPT-5.6 Terra in your tools
- Codex CLI setup — Use OpenAI Codex CLI and the Codex IDE extension with GPT-6 via a custom provider.
- DeepSeek Harness (dsh) setup with Claude & GPT — Add TokenMilo as a custom provider in DeepSeek’s open-source agent harness.
- WorkBuddy custom model setup — Use Claude and GPT in Tencent’s WorkBuddy desktop agent via a custom API.
- CC Switch one-click setup (Claude Code & Codex) — The easiest way: import TokenMilo into Claude Code and Codex with one click.
- OpenCode setup — Add TokenMilo as a custom provider in the OpenCode terminal agent.
- OpenClaw setup — Run the OpenClaw personal agent on Claude and GPT via TokenMilo.
- Aider setup — Pair-program in your terminal with Claude or GPT via TokenMilo.
- Cursor setup — Add Claude and GPT models to Cursor with your own key.
- Cline setup (VS Code) — Configure the Cline coding agent with Claude via TokenMilo.
- Kilo Code setup — Add a TokenMilo custom provider to the Kilo Code agent.
- Roo Code setup (VS Code) — Use Claude 5.x and GPT-6 in Roo Code via the OpenAI Compatible provider.
- Continue setup (VS Code & JetBrains) — Configure Claude and GPT in Continue’s config.yaml.
- Trae custom model setup — Add Claude and GPT to the Trae IDE as custom models.
- Zed setup — Use TokenMilo models in the Zed editor’s agent panel.
- Chatbox setup — Chat with Claude and GPT in Chatbox on desktop and mobile.
- Cherry Studio setup — Chat with Claude and GPT in the Cherry Studio desktop app.
More OpenAI models
- GPT-6 Astra — $0.855 / $4.28
- GPT-6.1 Sol — $0.171 / $0.855
- GPT-6.1 Terra — $0.086 / $0.432
- GPT-6 Sol — $0.171 / $0.855
- GPT-6 Luna — $0.0081 / $0.043
Common GPT-5.6 Terra API errors and how to fix them
Most errors come from the key, the base URL or the model ID. Check these first:
401 Unauthorized / invalid API key
Cause: The key is wrong, has an extra space or line break, was deleted, or is an official Anthropic/OpenAI key instead of a TokenMilo key.
Fix: Copy the key again from the Tokens page of the console (it starts with sk-). In Claude Code, set it as ANTHROPIC_AUTH_TOKEN or ANTHROPIC_API_KEY — not both with different values.
The response is an HTML page, or "Unexpected token <" / JSON parse error
Cause: The base URL points at the website instead of the API, or has a typo.
Fix: Use https://api.tokenmilo.com/v1 as the base URL for OpenAI SDKs, Codex CLI and most tools. If a tool asks for a single URL, https://api.tokenmilo.com/v1 works everywhere.
model_not_found / "no available channel for model"
Cause: The model name is misspelled, has the wrong case, or includes a vendor prefix.
Fix: Use the exact model ID gpt-5.6-terra — lowercase, with no prefix such as anthropic/ or openai/.
Insufficient quota / balance
Cause: Your account balance is used up, or the key has its own spending limit that has been reached.
Fix: Top up in the console, or edit the key and raise or remove its quota limit.
Connection error, timeout, ECONNRESET or "fetch failed"
Cause: A local proxy (for example Clash in fake-ip mode) or a firewall is intercepting the connection.
Fix: Set api.tokenmilo.com to DIRECT in your proxy rules, or turn the proxy off. TokenMilo does not need a proxy.
429 Too Many Requests
Cause: Too many concurrent requests in a short time.
Fix: Retry with exponential backoff and lower the concurrency of batch jobs.
Context too long ("prompt is too long" / context_length_exceeded)
Cause: The conversation plus attached files exceeds the model’s context window.
Fix: Start a new session or run /compact in Claude Code, attach fewer files, and trim long tool outputs.
400 Unsupported parameter: max_tokens / temperature
Cause: Reasoning models reject some legacy Chat Completions parameters.
Fix: Use max_completion_tokens instead of max_tokens, remove temperature/top_p, or switch to the Responses API.
400 Invalid value for reasoning_effort
Cause: The tool sent a reasoning level the model doesn’t accept.
Fix: Use low, medium or high. In Codex CLI, set model_reasoning_effort in config.toml accordingly.
500 / 502 / 503 upstream error
Cause: A temporary error from the upstream model provider.
Fix: Retry after a few seconds. If it keeps failing for several minutes, contact support with the time and request ID.
Still stuck? Ask in our Telegram support group and we will help, including remote setup. Telegram →
FAQ
Is this the real GPT-5.6 Terra?
Yes. Requests with model ID gpt-5.6-terra are served by the genuine OpenAI model — no quantized, swapped or downgraded variant. Response quality is the same as calling OpenAI directly.
How much does GPT-5.6 Terra cost on TokenMilo?
$0.171 per 1M input tokens and $1.03 per 1M output tokens, compared with $2 / $12 officially — 91% less. There is no subscription or monthly minimum.
Which API format should I use for GPT-5.6 Terra?
Use the OpenAI Chat Completions or Responses API at https://api.tokenmilo.com/v1. Codex CLI and most OpenAI SDK-based tools work by changing only the base URL and key.
Does GPT-5.6 Terra support prompt caching?
Yes. Cache reads are billed at $0.017 per 1M tokens, which makes long system prompts and coding-agent sessions much cheaper.
Start using GPT-5.6 Terra
Sign up, top up, create a key — and point your SDK at TokenMilo.