sambanova support (#5547) (#5703)

* add sambanova support * sambanova support * updated api endpoint for sambanova --------- Co-authored-by: Venu Anuganti <venu@venublog.com> Co-authored-by: Venu Anuganti <venu@vairmac2020>
2024-09-14 17:23:04 -07:00 · 2024-09-14 17:23:04 -07:00 · 7c2ddba6c6
commit 7c2ddba6c6
parent 85acdb9193
7 changed files with 283 additions and 0 deletions
--- a/docs/my-website/docs/providers/sambanova.md
+++ b/docs/my-website/docs/providers/sambanova.md
@ -0,0 +1,143 @@
+import Tabs from '@theme/Tabs';
+import TabItem from '@theme/TabItem';
+
+# Sambanova
+https://community.sambanova.ai/t/create-chat-completion-api/
+
+:::tip
+
+**We support ALL Sambanova models, just set `model=sambanova/<any-model-on-sambanova>` as a prefix when sending litellm requests. For the complete supported model list, visit https://sambanova.ai/technology/models **
+
+:::
+
+## API Key
+```python
+# env variable
+os.environ['SAMBANOVA_API_KEY']
+```
+
+## Sample Usage
+```python
+from litellm import completion
+import os
+
+os.environ['SAMBANOVA_API_KEY'] = ""
+response = completion(
+    model="sambanova/Meta-Llama-3.1-8B-Instruct",
+    messages=[
+        {
+            "role": "user",
+            "content": "What do you know about sambanova.ai",
+        }
+    ],
+    max_tokens=10,
+    response_format={ "type": "json_object" },
+    seed=123,
+    stop=["\n\n"],
+    temperature=0.2,
+    top_p=0.9,
+    tool_choice="auto",
+    tools=[],
+    user="user",
+)
+print(response)
+```
+
+## Sample Usage - Streaming
+```python
+from litellm import completion
+import os
+
+os.environ['SAMBANOVA_API_KEY'] = ""
+response = completion(
+    model="sambanova/Meta-Llama-3.1-8B-Instruct",
+    messages=[
+        {
+            "role": "user",
+            "content": "What do you know about sambanova.ai",
+        }
+    ],
+    stream=True,
+    max_tokens=10,
+    response_format={ "type": "json_object" },
+    seed=123,
+    stop=["\n\n"],
+    temperature=0.2,
+    top_p=0.9,
+    tool_choice="auto",
+    tools=[],
+    user="user",
+)
+
+for chunk in response:
+    print(chunk)
+```
+
+
+## Usage with LiteLLM Proxy Server
+
+Here's how to call a Sambanova model with the LiteLLM Proxy Server
+
+1. Modify the config.yaml 
+
+  ```yaml
+  model_list:
+    - model_name: my-model
+      litellm_params:
+        model: sambanova/<your-model-name>  # add sambanova/ prefix to route as Sambanova provider
+        api_key: api-key                 # api key to send your model
+  ```
+
+
+2. Start the proxy 
+
+  ```bash
+  $ litellm --config /path/to/config.yaml
+  ```
+
+3. Send Request to LiteLLM Proxy Server
+
+  <Tabs>
+
+  <TabItem value="openai" label="OpenAI Python v1.0.0+">
+
+  ```python
+  import openai
+  client = openai.OpenAI(
+      api_key="sk-1234",             # pass litellm proxy key, if you're using virtual keys
+      base_url="http://0.0.0.0:4000" # litellm-proxy-base url
+  )
+
+  response = client.chat.completions.create(
+      model="my-model",
+      messages = [
+          {
+              "role": "user",
+              "content": "what llm are you"
+          }
+      ],
+  )
+
+  print(response)
+  ```
+  </TabItem>
+
+  <TabItem value="curl" label="curl">
+
+  ```shell
+  curl --location 'http://0.0.0.0:4000/chat/completions' \
+      --header 'Authorization: Bearer sk-1234' \
+      --header 'Content-Type: application/json' \
+      --data '{
+      "model": "my-model",
+      "messages": [
+          {
+          "role": "user",
+          "content": "what llm are you"
+          }
+      ],
+  }'
+  ```
+  </TabItem>
+
+  </Tabs>
--- a/docs/my-website/sidebars.js
+++ b/docs/my-website/sidebars.js
@ -174,6 +174,7 @@ const sidebars = {
        "providers/baseten", 
        "providers/openrouter", 
        "providers/palm", 
+        "providers/sambanova", 
        // "providers/custom_openai_proxy",
        "providers/custom_llm_server",
        "providers/petals",
--- a/litellm/init.py
+++ b/litellm/init.py
@ -463,6 +463,7 @@ openai_compatible_endpoints: List = [
    "api.together.xyz/v1",
    "app.empower.dev/api/v1",
    "inference.friendli.ai/v1",
+    "api.sambanova.ai/v1",
 ]

 # this is maintained for Exception Mapping
@ -472,6 +473,7 @@ openai_compatible_providers: List = [
    "groq",
    "nvidia_nim",
    "cerebras",
+    "sambanova",
    "ai21_chat",
    "volcengine",
    "codestral",
@ -724,6 +726,7 @@ class LlmProviders(str, Enum):
    CODESTRAL = "codestral"
    TEXT_COMPLETION_CODESTRAL = "text-completion-codestral"
    DEEPSEEK = "deepseek"
+    SAMBANOVA = "sambanova"
    MARITALK = "maritalk"
    VOYAGE = "voyage"
    CLOUDFLARE = "cloudflare"
@ -953,6 +956,7 @@ from .llms.OpenAI.gpt_transformation import (
 )
 from .llms.nvidia_nim import NvidiaNimConfig
 from .llms.cerebras.chat import CerebrasConfig
+from .llms.sambanova.chat import SambanovaConfig
 from .llms.AI21.chat import AI21ChatConfig
 from .llms.fireworks_ai import FireworksAIConfig
 from .llms.volcengine import VolcEngineConfig
--- a/litellm/litellm_core_utils/get_llm_provider_logic.py
+++ b/litellm/litellm_core_utils/get_llm_provider_logic.py
@ -173,6 +173,13 @@ def get_llm_provider(
                    or "https://api.cerebras.ai/v1"
                )  # type: ignore
                dynamic_api_key = api_key or get_secret("CEREBRAS_API_KEY")
+            elif custom_llm_provider == "sambanova":
+                api_base = (
+                    api_base
+                    or get_secret("SAMBANOVA_API_BASE")
+                    or "https://api.sambanova.ai/v1"
+                )  # type: ignore
+                dynamic_api_key = api_key or get_secret("SAMBANOVA_API_KEY")
            elif (custom_llm_provider == "ai21_chat") or (
                custom_llm_provider == "ai21" and model in litellm.ai21_chat_models
            ):
@ -336,6 +343,9 @@ def get_llm_provider(
                    elif endpoint == "https://api.cerebras.ai/v1":
                        custom_llm_provider = "cerebras"
                        dynamic_api_key = get_secret("CEREBRAS_API_KEY")
+                    elif endpoint == "https://api.sambanova.ai/v1":
+                        custom_llm_provider = "sambanova"
+                        dynamic_api_key = get_secret("SAMBANOVA_API_KEY")
                    elif endpoint == "https://api.ai21.com/studio/v1":
                        custom_llm_provider = "ai21_chat"
                        dynamic_api_key = get_secret("AI21_API_KEY")
--- a/litellm/llms/sambanova/chat.py
+++ b/litellm/llms/sambanova/chat.py
@ -0,0 +1,91 @@
+"""
+Sambanova Chat Completions API
+
+this is OpenAI compatible - no translation needed / occurs
+"""
+
+import types
+from typing import Optional
+
+
+class SambanovaConfig:
+    """
+    Reference: https://community.sambanova.ai/t/create-chat-completion-api/
+
+    Below are the parameters:
+    """
+
+    max_tokens: Optional[int] = None
+    response_format: Optional[dict] = None
+    seed: Optional[int] = None
+    stop: Optional[str] = None
+    stream: Optional[bool] = None
+    temperature: Optional[float] = None
+    top_p: Optional[int] = None
+    tool_choice: Optional[str] = None
+    tools: Optional[list] = None
+    user: Optional[str] = None
+
+    def __init__(
+        self,
+        max_tokens: Optional[int] = None,
+        response_format: Optional[dict] = None,
+        seed: Optional[int] = None,
+        stop: Optional[str] = None,
+        stream: Optional[bool] = None,
+        temperature: Optional[float] = None,
+        top_p: Optional[int] = None,
+        tool_choice: Optional[str] = None,
+        tools: Optional[list] = None,
+        user: Optional[str] = None,
+    ) -> None:
+        locals_ = locals().copy()
+        for key, value in locals_.items():
+            if key != "self" and value is not None:
+                setattr(self.__class__, key, value)
+
+    @classmethod
+    def get_config(cls):
+        return {
+            k: v
+            for k, v in cls.__dict__.items()
+            if not k.startswith("__")
+            and not isinstance(
+                v,
+                (
+                    types.FunctionType,
+                    types.BuiltinFunctionType,
+                    classmethod,
+                    staticmethod,
+                ),
+            )
+            and v is not None
+        }
+
+    def get_supported_openai_params(self, model: str) -> list:
+        """
+        Get the supported OpenAI params for the given model
+
+        """
+
+        return [
+            "max_tokens",
+            "response_format",
+            "seed",
+            "stop",
+            "stream",
+            "temperature",
+            "top_p",
+            "tool_choice",
+            "tools",
+            "user",
+        ]
+
+    def map_openai_params(
+        self, model: str, non_default_params: dict, optional_params: dict
+    ) -> dict:
+        supported_openai_params = self.get_supported_openai_params(model=model)
+        for param, value in non_default_params.items():
+            if param in supported_openai_params:
+                optional_params[param] = value
+        return optional_params
--- a/litellm/main.py
+++ b/litellm/main.py
@ -388,6 +388,7 @@ async def acompletion(
            or custom_llm_provider == "groq"
            or custom_llm_provider == "nvidia_nim"
            or custom_llm_provider == "cerebras"
+            or custom_llm_provider == "sambanova"
            or custom_llm_provider == "ai21_chat"
            or custom_llm_provider == "volcengine"
            or custom_llm_provider == "codestral"
@ -1310,6 +1311,7 @@ def completion(
            or custom_llm_provider == "groq"
            or custom_llm_provider == "nvidia_nim"
            or custom_llm_provider == "cerebras"
+            or custom_llm_provider == "sambanova"
            or custom_llm_provider == "ai21_chat"
            or custom_llm_provider == "volcengine"
            or custom_llm_provider == "codestral"
@ -3164,6 +3166,7 @@ async def aembedding(*args, **kwargs) -> EmbeddingResponse:
            or custom_llm_provider == "groq"
            or custom_llm_provider == "nvidia_nim"
            or custom_llm_provider == "cerebras"
+            or custom_llm_provider == "sambanova"
            or custom_llm_provider == "ai21_chat"
            or custom_llm_provider == "volcengine"
            or custom_llm_provider == "deepseek"
@ -3829,6 +3832,7 @@ async def atext_completion(
            or custom_llm_provider == "groq"
            or custom_llm_provider == "nvidia_nim"
            or custom_llm_provider == "cerebras"
+            or custom_llm_provider == "sambanova"
            or custom_llm_provider == "ai21_chat"
            or custom_llm_provider == "volcengine"
            or custom_llm_provider == "text-completion-codestral"
--- a/litellm/model_prices_and_context_window_backup.json
+++ b/litellm/model_prices_and_context_window_backup.json
@ -1392,6 +1392,36 @@
        "mode": "chat",
        "supports_function_calling": true
    },
+    "sambanova/Meta-Llama-3.1-8B-Instruct": {
+        "max_tokens": 256000,
+        "max_input_tokens": 256000,
+        "max_output_tokens": 256000,
+        "input_cost_per_token": 0.0000001,
+        "output_cost_per_token": 0.0000001,
+        "litellm_provider": "sambanova",
+        "mode": "chat",
+        "supports_function_calling": true
+    },
+    "sambanova/Meta-Llama-3.1-70B-Instruct": {
+        "max_tokens": 256000,
+        "max_input_tokens": 256000,
+        "max_output_tokens": 256000,
+        "input_cost_per_token": 0.0000003,
+        "output_cost_per_token": 0.0000003,
+        "litellm_provider": "sambanova",
+        "mode": "chat",
+        "supports_function_calling": true
+    },
+    "sambanova/Meta-Llama-3.1-405B-Instruct": {
+        "max_tokens": 256000,
+        "max_input_tokens": 256000,
+        "max_output_tokens": 256000,
+        "input_cost_per_token": 0.0000006,
+        "output_cost_per_token": 0.0000006,
+        "litellm_provider": "sambanova",
+        "mode": "chat",
+        "supports_function_calling": true
+    },
    "friendliai/mixtral-8x7b-instruct-v0-1": {
        "max_tokens": 32768,
        "max_input_tokens": 32768,