scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-10-03 02:22:24 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 140

fix: move Nanobridge live tests to llm_translation; remove in-repo docs

Browse source
This commit is contained in:
wangzengwei 2026-06-03 23:20:54 +08:00
parent e2bc9d9ec3
commit d067d1adeb
3 changed files with 32 additions and 116 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

93
docs/my-website/docs/providers/nanobridge.md
View file

@ -1,93 +0,0 @@
# Nanobridge
| | |
|---|---|
| **Provider** | `nanobridge` |
| **API key env** | `NANOBRIDGE_API_KEY` |
| **Base override env** | `NANOBRIDGE_API_BASE` (optional) |
Nanobridge is an OpenAI-compatible API gateway for Tencent Hunyuan model routes (DeepSeek, GLM, MiniMax, etc.). It also exposes an Anthropic Messages-compatible base for Claude Code and similar agents.
## Quick Start
```bash
export NANOBRIDGE_API_KEY="your-api-key"
```
```python
import litellm
response = litellm.completion(
model="nanobridge/deepseek-v4-flash",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)
```
## Regional endpoints
Three regions share the same API key and billing. Pick the closest node for lower latency.
| Region | OpenAI base (`/v1`) | Anthropic base |
|--------|---------------------|----------------|
| Germany (default) | `https://api.nanobridge.net/v1` | `https://api.nanobridge.net/anthropic` |
| Singapore | `https://api-sg.nanobridge.net/v1` | `https://api-sg.nanobridge.net/anthropic` |
| United States | `https://api-us.nanobridge.net/v1` | `https://api-us.nanobridge.net/anthropic` |
Override the default Germany URL:
```bash
export NANOBRIDGE_API_BASE="https://api-sg.nanobridge.net/v1"
```
## Supported models
Use `nanobridge/<model_id>` in LiteLLM:
| Model ID | Notes |
|----------|--------|
| `deepseek-v4-flash` | Recommended default |
| `deepseek-v4-pro` | |
| `deepseek-v3.2` | |
| `glm-5.1` | Alias: `glm-5-1` |
| `minimax-m2.7` | Alias: `minimax-m-2-7` |
## API paths
- Chat: `POST /chat/completions` or `POST /v1/chat/completions`
- Models: `GET /v1/models`
- Anthropic: `POST /v1/messages` on the Anthropic base
- Streaming: `stream: true` (SSE)
- Auth: `Authorization: Bearer <API_KEY>`
## LiteLLM Proxy
```yaml
model_list:
- model_name: nanobridge-deepseek-flash
litellm_params:
model: nanobridge/deepseek-v4-flash
api_key: os.environ/NANOBRIDGE_API_KEY
# api_base: os.environ/NANOBRIDGE_API_BASE
```
## OpenAI Python SDK
```python
from openai import OpenAI
import os
client = OpenAI(
api_key=os.environ["NANOBRIDGE_API_KEY"],
base_url="https://api.nanobridge.net/v1",
)
client.chat.completions.create(
model="deepseek-v4-flash",
messages=[{"role": "user", "content": "Hello"}],
)
```
## Links
- [Nanobridge API documentation](https://platform.nanobridge.net/#/api-docs)
- [Platform console](https://platform.nanobridge.net)

27
tests/llm_translation/test_nanobridge.py Normal file
View file

@ -0,0 +1,27 @@
"""
Integration tests for Nanobridge (real API calls).
Requires NANOBRIDGE_API_KEY. Not collected in mock-only CI for tests/test_litellm/.
"""
import os
import sys
import pytest
sys.path.insert(0, os.path.abspath("../.."))
import litellm
@pytest.mark.skipif(
not os.environ.get("NANOBRIDGE_API_KEY"),
reason="NANOBRIDGE_API_KEY not set",
)
def test_nanobridge_live_chat_completions():
response = litellm.completion(
model="nanobridge/deepseek-v4-flash",
messages=[{"role": "user", "content": "Reply with exactly: ok"}],
max_tokens=16,
)
assert response.choices[0].message.content

28
tests/test_litellm/llms/openai_like/test_nanobridge_provider.py
View file

@ -1,23 +1,19 @@
"""
Unit tests for Nanobridge JSON-configured OpenAI-compatible provider.
Mock-only unit tests for Nanobridge JSON-configured OpenAI-compatible provider.
Live / network tests: tests/llm_translation/test_nanobridge.py
"""
import os
import sys
try:
import pytest
except ImportError:
pytest = None
workspace_path = os.path.abspath(os.path.join(os.path.dirname(__file__), "../../../.."))
sys.path.insert(0, workspace_path)
import litellm
from litellm import LlmProviders
from litellm.llms.openai_like.json_loader import JSONProviderRegistry
from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
workspace_path = os.path.abspath(os.path.join(os.path.dirname(__file__), "../../../.."))
sys.path.insert(0, workspace_path)
class TestNanobridgeProvider:
def test_nanobridge_in_llm_providers_enum(self):
@ -60,17 +56,3 @@ class TestNanobridgeProvider:
mock_response="PONG",
)
assert response.choices[0].message.content == "PONG"
@pytest.mark.skipif(
not os.environ.get("NANOBRIDGE_API_KEY"),
reason="NANOBRIDGE_API_KEY not set",
)
class TestNanobridgeLive:
def test_nanobridge_live_chat_completions(self):
response = litellm.completion(
model="nanobridge/deepseek-v4-flash",
messages=[{"role": "user", "content": "Reply with exactly: ok"}],
max_tokens=16,
)
assert response.choices[0].message.content