# Newcclm

> Newcclm is an AI assistant service that can be used immediately after opening. Users can use the command line client by registering an account;
> users who need to access their own tools, they can access according to the OpenAI compatible protocol.

Key Facts

- Interface address (Base URL): https://sccclm.com/en/v1
- Protocol: OpenAI compatible (/v1/chat/completions). Anthropic native protocol is not supported.
- Authentication: `Authorization: Bearer <API_KEY>`
- Available model names: `newcclm-fast` (fast file), `newcclm-coding` (enhanced file)
- Do not fill in real model names such as `gpt-4o`, `claude-*`, `deepseek-*`, etc., it will report that the model does not exist.
- Supports streaming output (`"stream": true`)
- Context length: 1M token

# # Underlying model (public)

| underlying model | corresponding gear | purpose |
| --- | --- | --- |
| GLM-5.3-Flash | `newcclm` / `newcclm-fast` | Daily main |
| Qwen3.8-Flash | `newcclm-coding` | Encoding and long context |
| DeepSeek V4-Flash | Automatic failover | Backup line |

All three models are 1M contexts. Gears are not replaced by downgrades; DeepSeek only takes over if the main line fails.

**Users must also abide by the terms of use of each underlying model** (not used for illegal content, not resold model capabilities, etc.).


Price

- Personal profile: ¥30 / 30 days (Alipay), or $5 / 30 days (USDT TRC20 / Stripe international card)
- Includes $4 usage per month; can be replenished after use without direct interruption
- no automatic renewal
- Small team profile: starting from ¥300 / 30 days, including multiple accounts, invoices can be issued

Documents

- Help Center: https://sccclm.com/en/help/
- Client Guide: https://sccclm.com/en/help/client.html
- DeepSeek Harness configuration: https://sccclm.com/en/help/dsh.html
- FAQ: https://sccclm.com/en/help/faq.html
- Check usage: https://sccclm.com/en/account/
- Full version description (for AI to read): https://sccclm.com/en/llms-full.txt

# # Command line client

```
npx newcclm
```

The first run will prompt you to register or log in to your account. The client does not require the user to provide an API Key, nor does it need to configure any parameters.
Other commands: `--status` to view usage, `--login` to switch accounts, `--logout` to exit, `-p "question"` to ask a single question.

# # Common Error Comparison

| Returns | Meaning | Processes |
| --- | --- | --- |
| `Invalid token` | Wrong or invalid key | Check key; client user logs in again |
| `model not found` | The model name is not our stall name | You can only fill in `newcclm-fast` / `newcclm-coding` |
| `404` | Wrong Base URL | Should be `https://sccclm.com/en/v1`, do not superimpose `/v1` |
| The message contains `quota` | The current month's usage has been used up | Replenish the usage or wait for the next month |
| `429` | Trigger current limiting | Reduce concurrency and try again |
| `401` and the message contains `upstream` | Upstream failure, not user configuration issue | Try again later |

Description

Newcclm is an independent third-party tool service that is not affiliated with the vendor of the model being invoked.