scriptoriumadmin/litellm
1
0
Fork 0
mirror of https://github.com/BerriAI/litellm.git synced 2026-11-14 09:03:16 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 214

feat: add Opper as an OpenAI-compatible provider

Browse source 
Registers opper as a JSON-configured OpenAI-compatible provider, in the
same shape as the Reka entry: OPPER_API_KEY and OPPER_API_BASE, default
base https://api.opper.ai/v3/compat, native /v1/chat/completions,
/v1/responses and /v1/messages, the Admin UI Add Model form and the
endpoint support matrix. Pinned route ids such as
opper/anthropic/claude-sonnet-4-6 keep their provider segment.

Disclosure: Opper is my startup.
This commit is contained in:
Felix Wunderlich 2026-10-20 22:46:45 +02:00
parent d18efdfe13
commit e1ece4a6f0
7 changed files with 337 additions and 0 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
litellm/constants.py
View file

@ -1082,6 +1082,7 @@ openai_compatible_endpoints: Final[list[str]] = [
"https://gigachat.devices.sberbank.ru/api/v1",
"https://api.openinfer.ai/v1",
"https://vanchin.streamlake.ai/api/gateway/v1/endpoints",
"https://api.opper.ai/v3/compat",
]
@ -1165,6 +1166,7 @@ openai_compatible_providers: Final[list[str]] = [
"sail",
"reka",
"streamlake",
"opper",
]
OPENAI_AUDIO_TRANSCRIPTION_PROVIDERS: Final = frozenset({"openai"} | frozenset(openai_compatible_providers))

6
litellm/llms/openai_like/providers.json
View file

@ -292,5 +292,11 @@
"preserve_cache_control": true
},
"supported_endpoints": ["/v1/chat/completions"]
},
"opper": {
"base_url": "https://api.opper.ai/v3/compat",
"api_key_env": "OPPER_API_KEY",
"api_base_env": "OPPER_API_BASE",
"supported_endpoints": ["/v1/chat/completions", "/v1/responses", "/v1/messages"]
}
}

18
litellm/provider_endpoints_support_backup.json
View file

@ -3085,6 +3085,24 @@
"generateContent": false
}
},
"opper": {
"display_name": "Opper (`opper`)",
"url": "https://docs.litellm.ai/docs/providers/opper",
"endpoints": {
"chat_completions": true,
"messages": true,
"responses": true,
"embeddings": false,
"image_generations": false,
"audio_transcriptions": false,
"audio_speech": false,
"moderations": false,
"batches": false,
"rerank": false,
"a2a": false,
"interactions": false
}
},
"ovhcloud": {
"display_name": "OVHCloud AI Endpoints (`ovhcloud`)",
"url": "https://docs.litellm.ai/docs/providers/ovhcloud",

28
litellm/proxy/public_endpoints/provider_create_fields.json
View file

@ -3071,6 +3071,34 @@
],
"default_model_placeholder": "gpt-3.5-turbo"
},
{
"provider": "OPPER",
"provider_display_name": "Opper",
"litellm_provider": "opper",
"credential_fields": [
{
"key": "api_base",
"label": "API Base",
"placeholder": "https://api.opper.ai/v3/compat",
"tooltip": null,
"required": false,
"field_type": "text",
"options": null,
"default_value": null
},
{
"key": "api_key",
"label": "API Key",
"placeholder": null,
"tooltip": null,
"required": true,
"field_type": "password",
"options": null,
"default_value": null
}
],
"default_model_placeholder": "opper/claude-sonnet-4-6"
},
{
"provider": "Nadir",
"provider_display_name": "Nadir",

1
litellm/types/utils.py
View file

@ -4667,6 +4667,7 @@ class LlmProviders(str, Enum):
SAIL = "sail"
REKA = "reka"
STREAMLAKE = "streamlake"
OPPER = "opper"
LITELLM_AGENT = "litellm_agent"
CURSOR = "cursor"
BEDROCK_MANTLE = "bedrock_mantle"

18
provider_endpoints_support.json
View file

@ -3085,6 +3085,24 @@
"generateContent": false
}
},
"opper": {
"display_name": "Opper (`opper`)",
"url": "https://docs.litellm.ai/docs/providers/opper",
"endpoints": {
"chat_completions": true,
"messages": true,
"responses": true,
"embeddings": false,
"image_generations": false,
"audio_transcriptions": false,
"audio_speech": false,
"moderations": false,
"batches": false,
"rerank": false,
"a2a": false,
"interactions": false
}
},
"ovhcloud": {
"display_name": "OVHCloud AI Endpoints (`ovhcloud`)",
"url": "https://docs.litellm.ai/docs/providers/ovhcloud",

264
tests/unit/llms/openai_like/test_opper_provider.py Normal file
View file

@ -0,0 +1,264 @@
import json
from pathlib import Path
from typing import Final
import pytest
import respx
import litellm
from litellm.caching.llm_caching_handler import LLMClientCache
_CHAT_COMPLETION: Final = {
"id": "chatcmpl_opper",
"object": "chat.completion",
"created": 1_790_000_000,
"model": "claude-sonnet-4-6",
"choices": [
{
"index": 0,
"message": {"role": "assistant", "content": "Hello from Opper"},
"finish_reason": "stop",
}
],
"usage": {"prompt_tokens": 4, "completion_tokens": 3, "total_tokens": 7},
}
def test_opper_is_a_registered_provider():
assert litellm.LlmProviders.OPPER.value == "opper"
assert "opper" in litellm.provider_list
assert "opper" in litellm.constants.openai_compatible_providers
def test_opper_provider_resolution(monkeypatch: pytest.MonkeyPatch):
from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
monkeypatch.setenv("OPPER_API_KEY", "opper-test-key")
model, provider, api_key, api_base = get_llm_provider(
model="opper/claude-sonnet-4-6",
custom_llm_provider=None,
api_base=None,
api_key=None,
)
assert model == "claude-sonnet-4-6"
assert provider == "opper"
assert api_key == "opper-test-key"
assert api_base == "https://api.opper.ai/v3/compat"
def test_opper_pinned_route_keeps_its_provider_segment(monkeypatch: pytest.MonkeyPatch):
from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
monkeypatch.setenv("OPPER_API_KEY", "opper-test-key")
model, provider, _, _ = get_llm_provider(
model="opper/anthropic/claude-sonnet-4-6",
custom_llm_provider=None,
api_base=None,
api_key=None,
)
assert model == "anthropic/claude-sonnet-4-6"
assert provider == "opper"
def test_opper_provider_keeps_explicit_credentials(monkeypatch: pytest.MonkeyPatch):
from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
monkeypatch.setenv("OPPER_API_KEY", "opper-env-key")
monkeypatch.setenv("OPPER_API_BASE", "https://opper.env.example/v3/compat")
_, provider, api_key, api_base = get_llm_provider(
model="opper/claude-sonnet-4-6",
custom_llm_provider=None,
api_base="https://opper.internal.example/v3/compat",
api_key="opper-explicit-key",
)
assert provider == "opper"
assert api_key == "opper-explicit-key"
assert api_base == "https://opper.internal.example/v3/compat"
def test_opper_api_base_env_overrides_default(monkeypatch: pytest.MonkeyPatch):
from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
monkeypatch.setenv("OPPER_API_KEY", "opper-env-key")
monkeypatch.setenv("OPPER_API_BASE", "https://opper.env.example/v3/compat")
_, provider, _, api_base = get_llm_provider(
model="opper/claude-sonnet-4-6",
custom_llm_provider=None,
api_base=None,
api_key=None,
)
assert provider == "opper"
assert api_base == "https://opper.env.example/v3/compat"
def test_opper_api_base_autodetects_provider(monkeypatch: pytest.MonkeyPatch):
from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider
monkeypatch.setenv("OPPER_API_KEY", "opper-env-key")
model, provider, api_key, api_base = get_llm_provider(
model="claude-sonnet-4-6",
custom_llm_provider=None,
api_base="https://api.opper.ai/v3/compat",
api_key=None,
)
assert model == "claude-sonnet-4-6"
assert provider == "opper"
assert api_key == "opper-env-key"
assert api_base == "https://api.opper.ai/v3/compat"
def test_opper_is_available_in_add_model_form():
fields_path = Path(litellm.__file__).parent / "proxy" / "public_endpoints" / "provider_create_fields.json"
providers = json.loads(fields_path.read_text())
opper = next(provider for provider in providers if provider["litellm_provider"] == "opper")
assert opper["provider"] == "OPPER"
assert opper["provider_display_name"] == "Opper"
assert opper["default_model_placeholder"] == "opper/claude-sonnet-4-6"
assert {field["key"]: field["required"] for field in opper["credential_fields"]} == {
"api_base": False,
"api_key": True,
}
def test_opper_supported_endpoints():
expected: Final = {
"chat_completions": True,
"messages": True,
"responses": True,
"embeddings": False,
"image_generations": False,
"audio_transcriptions": False,
"audio_speech": False,
"moderations": False,
"batches": False,
"rerank": False,
"a2a": False,
"interactions": False,
}
backup_path = Path(litellm.__file__).parent / "provider_endpoints_support_backup.json"
root_path = Path(litellm.__file__).parent.parent / "provider_endpoints_support.json"
assert json.loads(backup_path.read_text())["providers"]["opper"]["endpoints"] == expected
assert json.loads(root_path.read_text())["providers"]["opper"]["endpoints"] == expected
def test_opper_chat_completion_request():
with respx.mock() as upstream:
route: Final = upstream.post("https://api.opper.ai/v3/compat/chat/completions").respond(
200, json=_CHAT_COMPLETION
)
response: Final = litellm.completion(
model="opper/claude-sonnet-4-6",
messages=[{"role": "user", "content": "Say hello"}],
api_key="opper-test-key",
)
request: Final = route.calls.last.request
body: Final = json.loads(request.content)
assert route.call_count == 1
assert str(request.url) == "https://api.opper.ai/v3/compat/chat/completions"
assert request.headers["authorization"] == "Bearer opper-test-key"
assert body["model"] == "claude-sonnet-4-6"
assert body["messages"] == [{"role": "user", "content": "Say hello"}]
assert response.choices[0].message.content == "Hello from Opper"
def test_opper_chat_completion_sends_pinned_route_unchanged():
with respx.mock() as upstream:
route: Final = upstream.post("https://api.opper.ai/v3/compat/chat/completions").respond(
200, json=_CHAT_COMPLETION
)
litellm.completion(
model="opper/anthropic/claude-sonnet-4-6",
messages=[{"role": "user", "content": "Say hello"}],
api_key="opper-test-key",
)
body: Final = json.loads(route.calls.last.request.content)
assert route.call_count == 1
assert body["model"] == "anthropic/claude-sonnet-4-6"
def test_opper_responses_request():
with respx.mock() as upstream:
route: Final = upstream.post("https://api.opper.ai/v3/compat/responses").respond(
200,
json={
"id": "resp_opper",
"object": "response",
"created_at": 1_790_000_000,
"model": "claude-sonnet-4-6",
"status": "completed",
"output": [
{
"id": "msg_opper",
"type": "message",
"role": "assistant",
"status": "completed",
"content": [{"type": "output_text", "text": "Hello from Opper", "annotations": []}],
}
],
"usage": {"input_tokens": 4, "output_tokens": 3, "total_tokens": 7},
},
)
response: Final = litellm.responses(
model="opper/claude-sonnet-4-6",
input="Say hello",
api_key="opper-test-key",
)
request: Final = route.calls.last.request
body: Final = json.loads(request.content)
assert route.call_count == 1
assert str(request.url) == "https://api.opper.ai/v3/compat/responses"
assert request.headers["authorization"] == "Bearer opper-test-key"
assert body["model"] == "claude-sonnet-4-6"
assert body["input"] == "Say hello"
assert response.output[0].content[0].text == "Hello from Opper"
@pytest.mark.asyncio
async def test_opper_anthropic_messages_request(monkeypatch: pytest.MonkeyPatch):
monkeypatch.setattr(litellm, "disable_aiohttp_transport", True)
monkeypatch.setattr(litellm, "in_memory_llm_clients_cache", LLMClientCache())
with respx.mock() as upstream:
route: Final = upstream.post("https://api.opper.ai/v3/compat/v1/messages").respond(
200,
json={
"id": "msg_opper",
"type": "message",
"role": "assistant",
"model": "claude-sonnet-4-6",
"content": [{"type": "text", "text": "Hello from Opper"}],
"stop_reason": "end_turn",
"stop_sequence": None,
"usage": {"input_tokens": 4, "output_tokens": 3},
},
)
response: Final = await litellm.anthropic.messages.acreate(
model="opper/claude-sonnet-4-6",
messages=[{"role": "user", "content": "Say hello"}],
max_tokens=32,
api_key="opper-test-key",
)
request: Final = route.calls.last.request
body: Final = json.loads(request.content)
assert route.call_count == 1
assert str(request.url) == "https://api.opper.ai/v3/compat/v1/messages"
assert request.headers["authorization"] == "Bearer opper-test-key"
assert request.headers["anthropic-version"] == "2023-06-01"
assert body["model"] == "claude-sonnet-4-6"
assert body["messages"] == [{"role": "user", "content": "Say hello"}]
assert response["content"][0]["text"] == "Hello from Opper"