From 280ad6f049fd25db972080cc55cf54d7779c79a3 Mon Sep 17 00:00:00 2001
From: Krrish Dholakia <krrishdholakia@gmail.com>
Date: Tue, 12 Aug 2025 17:17:19 -0700
Subject: [PATCH] fix(main.py): add sambanova api base support to embeddings

---
 litellm/main.py                                     | 10 ++++++++--
 litellm/model_prices_and_context_window_backup.json |  9 ---------
 2 files changed, 8 insertions(+), 11 deletions(-)

diff --git a/litellm/main.py b/litellm/main.py
index 754b824e7dd..339d9e14406 100644
--- a/litellm/main.py
+++ b/litellm/main.py
@@ -1890,7 +1890,7 @@ def completion(  # type: ignore # noqa: PLR0915
                 or get_secret_str("COMETAPI_KEY")
                 or litellm.api_key
             )
-            
+
             api_base = (
                 api_base
                 or litellm.api_base
@@ -1917,7 +1917,7 @@ def completion(  # type: ignore # noqa: PLR0915
                 stream=stream,
                 provider_config=provider_config,
             )
-            
+
             ## LOGGING
             logging.post_call(
                 input=messages, api_key=api_key, original_response=response
@@ -4281,6 +4281,12 @@ def embedding(  # noqa: PLR0915
             )
         elif custom_llm_provider == "sambanova":
             api_key = api_key or litellm.api_key or get_secret_str("SAMBANOVA_API_KEY")
+            api_base = (
+                api_base
+                or litellm.api_base
+                or get_secret_str("SAMBANOVA_API_BASE")
+                or "https://api.sambanova.ai/v1"
+            )
             response = base_llm_http_handler.embedding(
                 model=model,
                 input=input,
diff --git a/litellm/model_prices_and_context_window_backup.json b/litellm/model_prices_and_context_window_backup.json
index cf1e7590906..1001aca9c09 100644
--- a/litellm/model_prices_and_context_window_backup.json
+++ b/litellm/model_prices_and_context_window_backup.json
@@ -16902,15 +16902,6 @@
         "supports_reasoning": true,
         "source": "https://cloud.sambanova.ai/plans/pricing"
     },
-    "sambanova/E5-Mistral-7B-Instruct": {
-        "max_tokens": 4096,
-        "max_input_tokens": 4096,
-        "input_cost_per_token": 1.3e-07,
-        "output_cost_per_token": 0.0,
-        "litellm_provider": "sambanova-embedding-models",
-        "mode": "embedding",
-        "source": "https://cloud.sambanova.ai/plans/pricing"
-    },
     "assemblyai/nano": {
         "mode": "audio_transcription",
         "input_cost_per_second": 0.00010278,