git clone https://github.com/diyism/cc-nim
cd cc-nim
cp .env.exmpale .env
#设置 nvidia key
nano .env
./cc-nim.sh &

#测试流式请求, 快速持续返回很多行就说明成功启动了:
curl -N -X POST http://localhost:3001/v1/messages     -H "Content-Type: application/json"     -H "x-api-key: test"     -d '{
     "model": "z-ai/glm4.7",
     "max_tokens": 256,
     "messages": [{"role": "user", "content": "Say hello in one sentence"}],
     "stream": true
   }'

cc-nim

Use Claude Code CLI for free with NVIDIA NIM's free unlimited 40 reqs/min API. This lightweight proxy converts Claude Code's Anthropic API requests to NVIDIA NIM format. Includes Telegram bot integration for remote control from your phone!

Quick Start

1. Get Your Free NVIDIA API Key

1. Get a new API key from build.nvidia.com/settings/api-keys
2. Install claude-code
3. Install uv

2. Clone & Configure

git clone https://github.com/Alishahryar1/cc-nim.git
cd cc-nim

cp .env.example .env

Edit .env:

NVIDIA_NIM_API_KEY=nvapi-your-key-here
MODEL=moonshotai/kimi-k2-thinking

---

Claude Code

Terminal 1 - Start the proxy:

uv run uvicorn server:app --host 0.0.0.0 --port 8082

Terminal 2 - Run Claude Code:

ANTHROPIC_BASE_URL=http://localhost:8082 claude

That's it! Claude Code now uses NVIDIA NIM for free.

---

Telegram Bot Integration

Control Claude Code remotely via Telegram! Send tasks from your phone and watch Claude work.

Setup

1. Get a Bot Token:

   • Open Telegram and message @BotFather
   • Send /newbot and follow the prompts
   • Copy the HTTP API Token

2. Add to .env:

TELEGRAM_BOT_TOKEN=123456789:ABCdefGHIjklMNOpqrSTUvwxYZ
ALLOWED_TELEGRAM_USER_ID=your_telegram_user_id

💡 To find your Telegram user ID, message @userinfobot on Telegram.

3. Configure the workspace (where Claude will operate):

CLAUDE_WORKSPACE=./agent_workspace
ALLOWED_DIR=C:/Users/yourname/projects

4. Start the server:

uv run uvicorn server:app --host 0.0.0.0 --port 8082

5. Usage:
   • Send /start to your bot
   • Send any text prompt to start a task

• Send a message to yourself on Telegram with a task
• Claude will respond with:
  • 💭 Thinking tokens (reasoning steps)
  • 🔧 Tool calls as they execute
  • ✅ Final result when complete
• Send /stop to cancel a running task

Available Models

See nvidia_nim_models.json for the full list of supported models.

Popular choices:

• moonshotai/kimi-k2.5
• z-ai/glm4.7
• minimaxai/minimax-m2.1
• mistralai/devstral-2-123b-instruct-2512

Browse all models at build.nvidia.com

Updating the Model List

To update nvidia_nim_models.json with the latest models from NVIDIA NIM, run the following command:

curl "https://integrate.api.nvidia.com/v1/models" > nvidia_nim_models.json

Configuration

Variable	Description	Default
NVIDIA_NIM_API_KEY	Your NVIDIA API key	required
MODEL	Model to use for all requests	moonshotai/kimi-k2-thinking
NVIDIA_NIM_BASE_URL	NIM endpoint	https://integrate.api.nvidia.com/v1
CLAUDE_WORKSPACE	Directory for agent workspace	./agent_workspace
ALLOWED_DIR	Allowed directories for agent	""
MAX_CLI_SESSIONS	Max concurrent CLI sessions	10
TELEGRAM_BOT_TOKEN	Telegram Bot Token	""
ALLOWED_TELEGRAM_USER_ID	Allowed Telegram User ID	""
MESSAGING_RATE_LIMIT	Telegram messages per window	1
MESSAGING_RATE_WINDOW	Messaging window (seconds)	1
NVIDIA_NIM_RATE_LIMIT	API requests per window	40
NVIDIA_NIM_RATE_WINDOW	Rate limit window (seconds)	60
NVIDIA_NIM_TEMPERATURE	Model temperature	1.0
NVIDIA_NIM_MAX_TOKENS	Max tokens for generation	81920

See .env.example for all supported parameters.

Development

Running Tests

To run the test suite, use the following command:

uv run pytest

Adding Your Own Provider

Extend BaseProvider in providers/ to add support for other APIs:

from providers.base import BaseProvider, ProviderConfig

class MyProvider(BaseProvider):
    async def complete(self, request):
        # Make API call, return raw JSON
        pass

    async def stream_response(self, request, input_tokens=0):
        # Yield Anthropic SSE format events
        pass

    def convert_response(self, response_json, original_request):
        # Convert to Anthropic response format
        pass

Adding Your Own Messaging App

Extend MessagingPlatform in messaging/ to add support for other platforms (Discord, Slack, etc.):

from messaging.base import MessagingPlatform
from messaging.models import IncomingMessage

class MyPlatform(MessagingPlatform):
    async def start(self):
        # Initialize connection
        pass

    async def stop(self):
        # Cleanup
        pass

    async def queue_send_message(self, chat_id, text, **kwargs):
        # Send message to platform
        pass

    async def queue_edit_message(self, chat_id, message_id, text, **kwargs):
        # Edit existing message
        pass

    def on_message(self, handler):
        # Register callback for incoming messages
        # Handler expects an IncomingMessage object
        pass