Short answer
Claude Sonnet 5.5 is available on TeamoRouter as claude-sonnet-5-5, with a 1,000,000-token context window and up to 128,000 output tokens. It costs about 30–35% of Anthropic's list price (list is $2 / $10 per million input / output tokens, unchanged from Sonnet 5); live prices are on the pricing page. In Claude Code it belongs in the default slot: set ANTHROPIC_MODEL and ANTHROPIC_DEFAULT_SONNET_MODEL to claude-sonnet-5-5, keep Opus 5.5 one /model command away for the hard tasks. You pay per token, by card or USDT, with no subscription.
Model ID, list price and curl examples: Claude Sonnet 5.5 API.
October 2026: Anthropic suspended many Claude accounts used over a VPN from unsupported regions such as Russia and Hong Kong (SCMP). This setup doesn't use an Anthropic account, so those suspensions don't affect it. If your account was suspended, see Claude Code banned: what to do.
No card at all? You can also top up with USDT, USDC or ETH: click Pay with crypto in the top-up dialog and send it from Binance, OKX or any wallet. The minimum is $5 and no browser extension is needed. Step by step: Pay for Claude and GPT APIs with crypto (USDT).
What Sonnet 5.5 is and when to use it
Sonnet is Anthropic's middle line — the one built for speed and price rather than maximum depth. Anthropic released Sonnet 5.5 on September 28, 2026 and describes it as the best combination of speed and intelligence in its current lineup. Its own positioning is clear: Opus 5.5 is for complex work that needs careful judgment; Sonnet 5.5 is strongest at well-scoped everyday tasks, fixing bugs, and producing polished documents.
In practice that makes it the model you leave running all day in Claude Code: bug fixes, features with a clear spec, tests, code review, documentation. Anthropic reports that it writes output more than 30% faster than Sonnet 5 and, because it needs fewer tokens and tool calls per task, costs up to 30% less per task at the same per-token price.
Sonnet 5.5 vs Opus 5.5 vs Sonnet 5
| Sonnet 5.5 | Opus 5.5 | Sonnet 5 | |
|---|---|---|---|
| Model ID | claude-sonnet-5-5 |
claude-opus-5-5 |
claude-sonnet-5 |
| Released | September 28, 2026 | September 22, 2026 | June 30, 2026 |
| List price, input / output per 1M | $2 / $10 | $4 / $20 | $2 / $10 |
| TeamoRouter | ≈30–35% of list | ≈30–35% of list | ≈30–35% of list |
| Context / max output | 1M / 128k | 1M / 128k | 1M / 128k |
| Anthropic's positioning | Best balance of speed and intelligence | Long-running agentic coding and knowledge work | Previous Sonnet, now legacy |
Sonnet 5 → Sonnet 5.5. Same input and output price, half the cache-read price ($0.10 instead of $0.20 per million at list), faster output, fewer tokens per task. Unless something in your pipeline depends on Sonnet 5's exact behaviour, there is no reason to stay on it — change the model name and check the thinking setting (see the FAQ below).
Sonnet 5.5 or Opus 5.5. Opus costs twice as much per token. Anthropic's own split: Opus 5.5 for complex work that needs careful judgment, Sonnet 5.5 for well-scoped everyday tasks. The practical rule: Sonnet 5.5 by default, Opus 5.5 when the task is open-ended and a wrong turn is expensive — a large refactor, an unfamiliar codebase, a design decision. For a cross-vendor view, see GPT-6 Sol vs Claude Opus 5.5.
Sonnet 5.5 API pricing per million tokens
List prices per 1 million tokens, in USD; the TeamoRouter column shows our price relative to list:
| Anthropic list price | TeamoRouter | |
|---|---|---|
| Input | $2.00 | ≈30–35% of list |
| Output | $10.00 | ≈30–35% of list |
| Cache read | $0.10 | ≈30–35% of list |
| Cache write | $2.50 | ≈30–35% of list |
Our exact price moves during the day, so live prices are always on the pricing page.
In a real session: a mid-size project in Claude Code reads about 200k input tokens and writes about 20k. At list price that is $0.40 + $0.20 = $0.60; on TeamoRouter it is about $0.18–0.21. The same session on Opus 5.5 costs $1.20 at list and about $0.36–0.42 through us — which is why Sonnet makes sense as the default and Opus as the exception. Repeated context is billed at the cache-read rate, so long sessions get cheaper still; see the billing FAQ.
Make Sonnet 5.5 the default in Claude Code
You need a TeamoRouter API key — create one in the console. If Claude Code isn't installed yet, start with the installation guide for macOS, Windows, and Linux.
macOS and Linux
Add to ~/.zshrc (or ~/.bashrc if you use bash):
export ANTHROPIC_BASE_URL=https://api.teamorouter.ai
export ANTHROPIC_AUTH_TOKEN="your TeamoRouter key"
export ANTHROPIC_MODEL=claude-sonnet-5-5
export ANTHROPIC_DEFAULT_SONNET_MODEL=claude-sonnet-5-5
export ANTHROPIC_DEFAULT_OPUS_MODEL=claude-opus-5-5
export ANTHROPIC_DEFAULT_HAIKU_MODEL=claude-haiku-4-5
Then source ~/.zshrc and run claude.
Windows (PowerShell)
setx ANTHROPIC_BASE_URL "https://api.teamorouter.ai"
setx ANTHROPIC_AUTH_TOKEN "your TeamoRouter key"
setx ANTHROPIC_MODEL "claude-sonnet-5-5"
setx ANTHROPIC_DEFAULT_SONNET_MODEL "claude-sonnet-5-5"
setx ANTHROPIC_DEFAULT_OPUS_MODEL "claude-opus-5-5"
setx ANTHROPIC_DEFAULT_HAIKU_MODEL "claude-haiku-4-5"
setx writes the variables permanently; close and reopen the terminal for them to take effect.
What each line does
ANTHROPIC_MODEL— the model a new session starts on. Here: Sonnet 5.5.ANTHROPIC_DEFAULT_SONNET_MODEL/..._OPUS_MODEL— what thesonnetandopusentries in/modelresolve to. Pinning them means that switching up to Opus lands on Opus 5.5, and switching back lands on Sonnet 5.5, not an older model.ANTHROPIC_DEFAULT_HAIKU_MODEL— the small model Claude Code uses for background chores.
Verify and switch inside a session
In a running Claude Code session type /model. The status line at the bottom should show Sonnet; the list should offer sonnet (→ claude-sonnet-5-5) and opus (→ claude-opus-5-5). Switch to opus for a hard refactor, back to sonnet when it's done.
To pin the setting for one project only, put it in .claude/settings.json at the repository root:
{
"env": {
"ANTHROPIC_BASE_URL": "https://api.teamorouter.ai",
"ANTHROPIC_MODEL": "claude-sonnet-5-5",
"ANTHROPIC_DEFAULT_SONNET_MODEL": "claude-sonnet-5-5"
}
}
Calling the API directly
The same key works outside Claude Code — from scripts, a backend, any client.
Anthropic format (what the official SDK uses):
curl https://api.teamorouter.ai/v1/messages \
-H "x-api-key: YOUR_KEY" \
-H "anthropic-version: 2023-06-01" \
-H "content-type: application/json" \
-d '{
"model": "claude-sonnet-5-5",
"max_tokens": 1024,
"messages": [{"role": "user", "content": "Find the bug in this function: ..."}]
}'
OpenAI-compatible format — if your code is written against the OpenAI SDK, only the base URL, key, and model name change:
from openai import OpenAI
client = OpenAI(base_url="https://api.teamorouter.ai/v1", api_key="YOUR_KEY")
resp = client.chat.completions.create(
model="claude-sonnet-5-5",
messages=[{"role": "user", "content": "Write a test for parse_date"}],
)
print(resp.choices[0].message.content)
Full parameter reference: API integration.
Paying for usage
You top up a balance in the console and are charged only for tokens actually used — no subscription, no monthly fee.
- Card — Visa, Mastercard and other major cards via Stripe.
- USDT — click Pay with crypto in the top-up dialog and pay with USDT, USDC or ETH from any exchange app; the checkout shows which networks your amount supports (BEP20 works from $5). No wallet extension needed. If your USDT is already on X Layer, install and unlock the OKX Wallet extension in a desktop browser and click Have USDT on X Layer? Pay with OKX Wallet in the top-up dialog. You only sign an authorization, so there is no gas fee. See how to pay with USDT.
There is no minimum bundle and nothing expires: top up $10 and you spend exactly $10 on tokens at the prices above.
FAQ
What is the Claude Sonnet 5.5 model ID? claude-sonnet-5-5. Use it as model in API requests; in Claude Code put it in ANTHROPIC_MODEL and ANTHROPIC_DEFAULT_SONNET_MODEL.
Claude Code warns that claude-sonnet-5-5 "isn't described by this version's model catalog". Your Claude Code is older than the model. Version 2.1.284 (September 28, 2026) is the first that knows Sonnet 5.5; older versions still work but assume a 200k context and compact the session early. Update with claude update (or npm install -g @anthropic-ai/claude-code@latest) and check with claude --version.
Is claude-sonnet-5-5 the same as claude-sonnet-5? No — two models, two IDs. Claude Code won't move you to 5.5 on its own; if your settings pin claude-sonnet-5, replace it.
Why does Sonnet 5.5 feel less thorough in Claude Code than in my API calls? The default effort level differs: Anthropic set it to medium in Claude Code and to high on the API. For a task that needs more thinking, raise the effort in Claude Code or switch to Opus 5.5.
My API code runs Sonnet with thinking turned off. Does it still work? Anthropic's migration note for Sonnet 5.5 says that if you run Sonnet with thinking off, you need to switch to the new between_tools setting, which keeps up-front thinking off. Check your request parameters before changing the model name in production.
How do I enable the 1M context? Nothing to enable — it's a property of the model. Every input token is billed, so on large repositories the cache matters more than the limit.
Next steps
- Claude Fable 5.5 API: pricing and setup — Anthropic's top-tier model.
- Claude Opus 5.5: pricing and setup — for the tasks where Sonnet isn't enough.
- What Claude Code really costs — a breakdown on typical workloads.
- Create an API key — and run Sonnet 5.5 in five minutes.