diff --git a/litellm/constants.py b/litellm/constants.py index 97631b9997a..ec6a00a29a0 100644 --- a/litellm/constants.py +++ b/litellm/constants.py @@ -1089,6 +1089,7 @@ openai_compatible_endpoints: Final[list[str]] = [ "https://api.cognition.ai/v1", "https://inference.coralbricks.ai/v1", "https://api.cortecs.ai/v1", + "https://llm.kimchi.dev/openai/v1", "https://api.scx.ai/v1", "https://api.prisminference.com/v1", "https://api.reka.ai/v1", @@ -1177,6 +1178,7 @@ openai_compatible_providers: Final[list[str]] = [ "prism", "hostinger_router", "sail", + "kimchi", # Kimchi - JSON-configured provider "reka", "streamlake", ] diff --git a/litellm/llms/openai_like/providers.json b/litellm/llms/openai_like/providers.json index a71a9c3bb9b..25c11a78343 100644 --- a/litellm/llms/openai_like/providers.json +++ b/litellm/llms/openai_like/providers.json @@ -302,5 +302,13 @@ "preserve_cache_control": true }, "supported_endpoints": ["/v1/chat/completions"] + }, + "kimchi": { + "base_url": "https://llm.kimchi.dev/openai/v1", + "api_key_env": "KIMCHI_API_KEY", + "api_base_env": "KIMCHI_API_BASE", + "supported_endpoints": [ + "/v1/chat/completions" + ] } } diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json index 77ce188ad57..9a793a68291 100644 --- a/litellm/model_prices_and_context_window_backup.json +++ b/litellm/model_prices_and_context_window_backup.json @@ -63904,6 +63904,137 @@ "supports_prompt_caching": true, "source": "https://docs.devin.ai/desktop/models" }, + "kimchi/kimi-k3": { + "max_tokens": 262144, + "max_input_tokens": 262144, + "max_output_tokens": 262144, + "input_cost_per_token": 3e-06, + "output_cost_per_token": 1.5e-05, + "cache_read_input_token_cost": 3e-07, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_vision": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, + "kimchi/kimi-k2.7": { + "max_tokens": 262144, + "max_input_tokens": 262144, + "max_output_tokens": 262144, + "input_cost_per_token": 9.5e-07, + "output_cost_per_token": 4e-06, + "cache_read_input_token_cost": 1.9e-07, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_vision": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, + "kimchi/nemotron-3-ultra-fp4": { + "max_tokens": 1000000, + "max_input_tokens": 1000000, + "max_output_tokens": 1000000, + "input_cost_per_token": 6e-07, + "output_cost_per_token": 3.6e-06, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_native_streaming": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, + "kimchi/deepseek-v4-flash": { + "max_tokens": 1000000, + "max_input_tokens": 1000000, + "max_output_tokens": 1000000, + "input_cost_per_token": 1.4e-07, + "output_cost_per_token": 2.8e-07, + "cache_read_input_token_cost": 7e-08, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_native_streaming": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, + "kimchi/deepseek-v4-flash-0731": { + "max_tokens": 1000000, + "max_input_tokens": 1000000, + "max_output_tokens": 1000000, + "input_cost_per_token": 1.4e-07, + "output_cost_per_token": 2.8e-07, + "cache_read_input_token_cost": 7e-08, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_native_streaming": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, + "kimchi/deepseek-v4.1-flash": { + "max_tokens": 1000000, + "max_input_tokens": 1000000, + "max_output_tokens": 1000000, + "input_cost_per_token": 3e-07, + "output_cost_per_token": 1.2e-06, + "cache_read_input_token_cost": 6e-09, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_native_streaming": true, + "supports_vision": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, + "kimchi/glm-5.3": { + "max_tokens": 1000000, + "max_input_tokens": 1000000, + "max_output_tokens": 1000000, + "input_cost_per_token": 1.4e-06, + "output_cost_per_token": 4.4e-06, + "cache_read_input_token_cost": 2.6e-07, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, + "kimchi/glm-5.3-flash": { + "max_tokens": 1000000, + "max_input_tokens": 1000000, + "max_output_tokens": 1000000, + "input_cost_per_token": 1.5e-07, + "output_cost_per_token": 5e-07, + "cache_read_input_token_cost": 3e-08, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_vision": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, + "kimchi/minimax-m3": { + "max_tokens": 1000000, + "max_input_tokens": 1000000, + "max_output_tokens": 1000000, + "input_cost_per_token": 3e-07, + "output_cost_per_token": 1.2e-06, + "cache_read_input_token_cost": 6e-08, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_parallel_function_calling": true, + "supports_prompt_caching": true, + "supports_response_schema": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_vision": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, "pinstripes/ps/glm-4.5-air": { "max_input_tokens": 128000, "max_output_tokens": 128000, diff --git a/litellm/proxy/public_endpoints/provider_create_fields.json b/litellm/proxy/public_endpoints/provider_create_fields.json index aa7554c4a5a..322215085fe 100644 --- a/litellm/proxy/public_endpoints/provider_create_fields.json +++ b/litellm/proxy/public_endpoints/provider_create_fields.json @@ -4452,5 +4452,33 @@ } ], "default_model_placeholder": "bigmodel/glm-5.3" + }, + { + "provider": "Kimchi", + "provider_display_name": "Kimchi", + "litellm_provider": "kimchi", + "credential_fields": [ + { + "key": "api_key", + "label": "Kimchi API Key", + "placeholder": null, + "tooltip": null, + "required": true, + "field_type": "password", + "options": null, + "default_value": null + }, + { + "key": "api_base", + "label": "API Base", + "placeholder": "https://llm.kimchi.dev/openai/v1", + "tooltip": null, + "required": false, + "field_type": "text", + "options": null, + "default_value": null + } + ], + "default_model_placeholder": "kimchi/kimi-k3" } ] diff --git a/litellm/types/utils.py b/litellm/types/utils.py index 57f4d971eb3..aa69ab651e8 100644 --- a/litellm/types/utils.py +++ b/litellm/types/utils.py @@ -4652,6 +4652,7 @@ class LlmProviders(str, Enum): V0 = "v0" MORPH = "morph" OPENCODE_GO = "opencode_go" + KIMCHI = "kimchi" LAMBDA_AI = "lambda_ai" INCEPTION = "inception" TEXT_COMPLETION_INCEPTION = "text-completion-inception" diff --git a/model_prices_and_context_window.json b/model_prices_and_context_window.json index 77ce188ad57..9a793a68291 100644 --- a/model_prices_and_context_window.json +++ b/model_prices_and_context_window.json @@ -63904,6 +63904,137 @@ "supports_prompt_caching": true, "source": "https://docs.devin.ai/desktop/models" }, + "kimchi/kimi-k3": { + "max_tokens": 262144, + "max_input_tokens": 262144, + "max_output_tokens": 262144, + "input_cost_per_token": 3e-06, + "output_cost_per_token": 1.5e-05, + "cache_read_input_token_cost": 3e-07, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_vision": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, + "kimchi/kimi-k2.7": { + "max_tokens": 262144, + "max_input_tokens": 262144, + "max_output_tokens": 262144, + "input_cost_per_token": 9.5e-07, + "output_cost_per_token": 4e-06, + "cache_read_input_token_cost": 1.9e-07, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_vision": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, + "kimchi/nemotron-3-ultra-fp4": { + "max_tokens": 1000000, + "max_input_tokens": 1000000, + "max_output_tokens": 1000000, + "input_cost_per_token": 6e-07, + "output_cost_per_token": 3.6e-06, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_native_streaming": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, + "kimchi/deepseek-v4-flash": { + "max_tokens": 1000000, + "max_input_tokens": 1000000, + "max_output_tokens": 1000000, + "input_cost_per_token": 1.4e-07, + "output_cost_per_token": 2.8e-07, + "cache_read_input_token_cost": 7e-08, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_native_streaming": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, + "kimchi/deepseek-v4-flash-0731": { + "max_tokens": 1000000, + "max_input_tokens": 1000000, + "max_output_tokens": 1000000, + "input_cost_per_token": 1.4e-07, + "output_cost_per_token": 2.8e-07, + "cache_read_input_token_cost": 7e-08, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_native_streaming": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, + "kimchi/deepseek-v4.1-flash": { + "max_tokens": 1000000, + "max_input_tokens": 1000000, + "max_output_tokens": 1000000, + "input_cost_per_token": 3e-07, + "output_cost_per_token": 1.2e-06, + "cache_read_input_token_cost": 6e-09, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_native_streaming": true, + "supports_vision": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, + "kimchi/glm-5.3": { + "max_tokens": 1000000, + "max_input_tokens": 1000000, + "max_output_tokens": 1000000, + "input_cost_per_token": 1.4e-06, + "output_cost_per_token": 4.4e-06, + "cache_read_input_token_cost": 2.6e-07, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, + "kimchi/glm-5.3-flash": { + "max_tokens": 1000000, + "max_input_tokens": 1000000, + "max_output_tokens": 1000000, + "input_cost_per_token": 1.5e-07, + "output_cost_per_token": 5e-07, + "cache_read_input_token_cost": 3e-08, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_prompt_caching": true, + "supports_reasoning": true, + "supports_vision": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, + "kimchi/minimax-m3": { + "max_tokens": 1000000, + "max_input_tokens": 1000000, + "max_output_tokens": 1000000, + "input_cost_per_token": 3e-07, + "output_cost_per_token": 1.2e-06, + "cache_read_input_token_cost": 6e-08, + "litellm_provider": "kimchi", + "mode": "chat", + "supports_function_calling": true, + "supports_parallel_function_calling": true, + "supports_prompt_caching": true, + "supports_response_schema": true, + "supports_system_messages": true, + "supports_tool_choice": true, + "supports_vision": true, + "source": "https://docs.kimchi.dev/docs/model-apis-pricing" + }, "pinstripes/ps/glm-4.5-air": { "max_input_tokens": 128000, "max_output_tokens": 128000, diff --git a/provider_endpoints_support.json b/provider_endpoints_support.json index 40ea9d15ff4..70272b4cb2c 100644 --- a/provider_endpoints_support.json +++ b/provider_endpoints_support.json @@ -2657,6 +2657,24 @@ "generateContent": false } }, + "kimchi": { + "display_name": "Kimchi (`kimchi`)", + "url": "https://docs.litellm.ai/docs/providers/kimchi", + "endpoints": { + "chat_completions": true, + "messages": false, + "responses": false, + "embeddings": false, + "image_generations": false, + "audio_transcriptions": false, + "audio_speech": false, + "moderations": false, + "batches": false, + "rerank": false, + "a2a": false, + "interactions": false + } + }, "morph": { "display_name": "Morph (`morph`)", "url": "https://docs.litellm.ai/docs/providers/morph", diff --git a/tests/unit/llms/openai_like/test_kimchi_provider.py b/tests/unit/llms/openai_like/test_kimchi_provider.py new file mode 100644 index 00000000000..053bfcbdf87 --- /dev/null +++ b/tests/unit/llms/openai_like/test_kimchi_provider.py @@ -0,0 +1,129 @@ +""" +Tests for the Kimchi provider identity. + +Kimchi exposes an OpenAI-compatible /v1/chat/completions surface at +https://llm.kimchi.dev/openai/v1. Its provider identity must resolve to a +distinct `kimchi` slug so OpenAI-specific pricing and provider-level reporting +never apply to traffic routed through Kimchi. +""" + +import json +from pathlib import Path +from typing import Final + +import pytest +import respx + +import litellm + + +class TestKimchiProviderIdentity: + def test_kimchi_is_a_registered_provider(self): + from litellm import LlmProviders + + assert LlmProviders.KIMCHI.value == "kimchi" + assert "kimchi" in litellm.provider_list + + def test_kimchi_json_config(self): + from litellm.llms.openai_like.json_loader import JSONProviderRegistry + + kimchi = JSONProviderRegistry.get("kimchi") + assert kimchi is not None + assert kimchi.base_url == "https://llm.kimchi.dev/openai/v1" + assert kimchi.api_key_env == "KIMCHI_API_KEY" + assert kimchi.api_base_env == "KIMCHI_API_BASE" + + def test_kimchi_in_openai_compatible_providers(self): + from litellm.constants import openai_compatible_providers + + assert "kimchi" in openai_compatible_providers + + def test_prefixed_model_resolves_to_kimchi_not_openai(self): + from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider + + model, provider, _, api_base = get_llm_provider( + model="kimchi/kimi-k3", + custom_llm_provider=None, + api_base=None, + api_key=None, + ) + + assert model == "kimi-k3" + assert provider == "kimchi" + assert api_base == "https://llm.kimchi.dev/openai/v1" + + def test_explicit_api_base_and_key_win(self): + from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider + + _, provider, api_key, api_base = get_llm_provider( + model="kimchi/kimi-k3", + custom_llm_provider=None, + api_base="https://kimchi.internal.example/v1", + api_key="sk-test", + ) + + assert provider == "kimchi" + assert api_base == "https://kimchi.internal.example/v1" + assert api_key == "sk-test" + + def test_api_base_autodetects_kimchi(self, monkeypatch): + from litellm.litellm_core_utils.get_llm_provider_logic import get_llm_provider + + monkeypatch.setenv("KIMCHI_API_KEY", "sk-test") + + _, provider, api_key, api_base = get_llm_provider( + model="kimi-k3", + custom_llm_provider=None, + api_base="https://llm.kimchi.dev/openai/v1", + api_key=None, + ) + + assert provider == "kimchi" + assert api_base == "https://llm.kimchi.dev/openai/v1" + assert api_key == "sk-test" + + +def test_kimchi_chat_request(monkeypatch: pytest.MonkeyPatch): + monkeypatch.setenv("KIMCHI_API_KEY", "kimchi-test-key") + + with respx.mock() as upstream: + route: Final = upstream.post("https://llm.kimchi.dev/openai/v1/chat/completions").respond( + 200, + json={ + "id": "chatcmpl-kimchi", + "object": "chat.completion", + "created": 1_789_550_000, + "model": "kimi-k3", + "choices": [ + { + "index": 0, + "finish_reason": "stop", + "message": {"role": "assistant", "content": "pong"}, + } + ], + "usage": {"prompt_tokens": 4, "completion_tokens": 1, "total_tokens": 5}, + }, + ) + response: Final = litellm.completion( + model="kimchi/kimi-k3", + messages=[{"role": "user", "content": "Reply with the single word: pong"}], + ) + + request: Final = route.calls.last.request + assert route.call_count == 1 + assert str(request.url) == "https://llm.kimchi.dev/openai/v1/chat/completions" + assert request.headers["authorization"] == "Bearer kimchi-test-key" + body: Final = json.loads(request.content) + assert body["model"] == "kimi-k3" + assert response.choices[0].message.content == "pong" + + +def test_kimchi_cost_map_registry_mirrors_cost_map(): + package_root: Final = Path(litellm.__file__).parent + + cost_map: Final = json.loads((package_root.parent / "model_prices_and_context_window.json").read_text()) + backup: Final = json.loads((package_root / "model_prices_and_context_window_backup.json").read_text()) + + kimchi_entries: Final = {name: cost_map[name] for name in cost_map if name.startswith("kimchi/")} + assert kimchi_entries + assert kimchi_entries == {name: backup[name] for name in kimchi_entries}