Claude Sonnet 5.5 is now live on GreenTokens at 50% off

Claude Sonnet 5.5 is available on GreenTokens from today. It succeeds Claude Sonnet 5 as Anthropic's everyday model for coding, agents and general work, with a 1 million token context window and up to 128,000 output tokens per response.

There's nothing to set up. If you already have a GreenTokens key, change the model in your request to claude-sonnet-5-5 and you're using it.

Prices

Claude Sonnet 5.5 is 50% below Anthropic's list price on every kind of token. Prices are per million tokens:

Anthropic kept Sonnet 5.5 at the same list price as Sonnet 5, so on GreenTokens the upgrade costs nothing extra per token. Current prices for every model are always on the models page.

Model ID and capabilities

• Model ID: claude-sonnet-5-5.
• Context window: 1 million tokens, with up to 128,000 output tokens per response.
• Input: text and images, so you can send screenshots, charts or photos with your prompt.
• Thinking: adaptive, with effort levels from low to max. The default is high.
• Prompt caching, streaming and tool use all work as they do on other Claude models.

Try it now

GreenTokens speaks the Anthropic Messages API, so the official Anthropic SDKs work unchanged. Point the client at https://api.greentokens.io with your GreenTokens key:

from anthropic import Anthropic

client = Anthropic(base_url="https://api.greentokens.io", api_key="sk-gt-…")

message = client.messages.create(
    model="claude-sonnet-5-5",
    max_tokens=16000,
    messages=[{"role": "user", "content": "Review this function for edge cases."}],
)
print(message.content[-1].text)

The text API docs have the same call in TypeScript and curl, plus streaming, tools, images and caching. Tools that only speak the OpenAI format can call Claude Sonnet 5.5 too, through https://api.greentokens.io/v1.

Compared with earlier Sonnet models

Prices are GreenTokens prices per million tokens. Sonnet 5.5 costs the same as Sonnet 5, and a third less than Sonnet 4.6. If you're still on Sonnet 4.6, moving to Sonnet 5.5 lowers your bill as well as upgrading the model.

What changed for developers

Most requests that work on Sonnet 5 work on Sonnet 5.5 as they are. Anthropic did change a few request options, which matter if your code sets them explicitly:

• Thinking can't be switched off with type "disabled" any more; that returns an error. To run without thinking, send type "between_tools", or keep thinking on at low effort.
• Forcing a tool with tool_choice "any" or a named tool returns an error. Use "auto" and name the tool in your prompt, or use structured outputs if you only need JSON back.
• Effort levels were recalibrated. The default is still high, but if you tuned effort for Sonnet 5, test again: medium is a good start for agentic coding, low for chat.

If you don't set any of these, there's nothing to change beyond the model ID.

Using Sonnet 5.5 in your tools

• Claude Code: set ANTHROPIC_DEFAULT_SONNET_MODEL to claude-sonnet-5-5, and the Sonnet choice in /model uses it.
• Claude Desktop: it reads the model list from GreenTokens, so Claude Sonnet 5.5 appears in the picker on its own.
• opencode: add claude-sonnet-5-5 to your GreenTokens provider's models, then pick it with /models.
• OpenClaw and Hermes Agent: use claude-sonnet-5-5 as the model ID in your GreenTokens provider.

Setting these tools up for the first time? The setup script connects them to GreenTokens in one command.

Switching safely

A new model can behave differently from the one you use today, even in the same family. A simple way to switch without surprises:

1. Create a separate API key for testing, with a low daily spend limit.
2. Send a representative sample of your real prompts to claude-sonnet-5-5 with that key.
3. Compare the answers with your current model, and check tokens and cost per request in the dashboard's Requests log.
4. Move traffic over gradually, starting with the requests where the new model did best.

Common questions

Do I need a new key or account?

No. Any existing GreenTokens key can call Claude Sonnet 5.5 straight away, billed from the same prepaid balance as every other model. If a key has spend limits, they apply to the new model too.

Is Claude Sonnet 5 still available?

Yes. claude-sonnet-5 keeps working at the same price, so you can compare the two side by side before you switch.

Is prompt caching supported?

Yes. Cache reads cost $0.10 per million tokens, a tenth of the input price, which makes a big difference for agents that resend the same long context on every turn. Read more in what prompt caching is.

Get started

New to GreenTokens? Sign in with your email, top up a prepaid balance, create an API key and follow the quickstart. There's no subscription, and Claude Sonnet 5.5 sits alongside Claude Opus, GPT, Grok, DeepSeek and the rest of the catalogue on one API.