feat(voyage): add voyage-context-4 and voyage-4 series models

docs/my-website/docs/providers/voyage.md

@@ -54,6 +54,9 @@ All models listed here https://docs.voyageai.com/embeddings/#models-and-specific
 
 | Model Name              | Function Call                                              |
 |-------------------------|------------------------------------------------------------|
+| voyage-4                | `embedding(model="voyage/voyage-4", input)`                | 
+| voyage-4-large          | `embedding(model="voyage/voyage-4-large", input)`          | 
+| voyage-4-lite           | `embedding(model="voyage/voyage-4-lite", input)`           | 
 | voyage-3.5              | `embedding(model="voyage/voyage-3.5", input)`              | 
 | voyage-3.5-lite         | `embedding(model="voyage/voyage-3.5-lite", input)`         | 
 | voyage-3-large          | `embedding(model="voyage/voyage-3-large", input)`          | 
@@ -72,9 +75,9 @@ All models listed here https://docs.voyageai.com/embeddings/#models-and-specific
 | voyage-lite-01          | `embedding(model="voyage/voyage-lite-01", input)`          |
 | voyage-lite-01-instruct | `embedding(model="voyage/voyage-lite-01-instruct", input)` |
 
-## Contextual Embeddings (voyage-context-3)
+## Contextual Embeddings (voyage-context-4)
 
-VoyageAI's `voyage-context-3` model provides contextualized chunk embeddings, where each chunk is embedded with awareness of its surrounding document context. This significantly improves retrieval quality compared to standard context-agnostic embeddings.
+VoyageAI's contextual models (`voyage-context-4`, and the previous generation `voyage-context-3`) provide contextualized chunk embeddings, where each chunk is embedded with awareness of its surrounding document context. This significantly improves retrieval quality compared to standard context-agnostic embeddings. `voyage-context-4` is the latest model and is recommended over `voyage-context-3`.
 
 ### Key Benefits
 - Chunks understand their position and role within the full document
@@ -94,7 +97,7 @@ os.environ['VOYAGE_API_KEY'] = "your-api-key"
 
 # Single document with multiple chunks
 response = embedding(
-    model="voyage/voyage-context-3",
+    model="voyage/voyage-context-4",
     input=[
         [
             "Chapter 1: Introduction to AI",
@@ -107,7 +110,7 @@ print(f"Number of chunk groups: {len(response.data)}")
 
 # Multiple documents
 response = embedding(
-    model="voyage/voyage-context-3",
+    model="voyage/voyage-context-4",
     input=[
         ["Paris is the capital of France.", "It is known for the Eiffel Tower."],
         ["Tokyo is the capital of Japan.", "It is a major economic hub."]
@@ -117,17 +120,17 @@ print(f"Processed {len(response.data)} documents")
 ```
 
 ### Specifications
-- Model: `voyage-context-3`
-- Context length: 32,000 tokens per document
+- Model: `voyage-context-4` (recommended), `voyage-context-3`
+- Context length: 32,000 tokens per chunk
 - Output dimensions: 256, 512, 1024 (default), or 2048
 - Max inputs: 1,000 per request
 - Max total tokens: 120,000
 - Max chunks: 16,000
-- Pricing: $0.18 per million tokens
+- Pricing: `voyage-context-4` $0.12 / `voyage-context-3` $0.18 per million tokens
 
 ### When to Use Contextual Embeddings
 
-**Use `voyage-context-3` when:**
+**Use `voyage-context-4` when:**
 - Processing long documents split into chunks
 - Document structure and flow are important
 - References between sections matter
@@ -143,13 +146,17 @@ print(f"Processed {len(response.data)} documents")
 
 | Model | Best For | Context Length | Price/M Tokens |
 |-------|----------|----------------|----------------|
+| voyage-4 | General-purpose, multilingual | 32K | $0.06 |
+| voyage-4-large | Best overall quality | 32K | $0.12 |
+| voyage-4-lite | Latency-sensitive applications | 32K | $0.02 |
 | voyage-3.5 | General-purpose, multilingual | 32K | $0.06 |
 | voyage-3.5-lite | Latency-sensitive applications | 32K | $0.02 |
 | voyage-3-large | Best overall quality | 32K | $0.18 |
 | voyage-code-3 | Code retrieval and search | 32K | $0.18 |
 | voyage-finance-2 | Financial documents | 32K | $0.12 |
 | voyage-law-2 | Legal documents | 16K | $0.12 |
-| voyage-context-3 | Contextual document embeddings | 32K | $0.18 |
+| voyage-context-4 | Contextual document embeddings (recommended) | 32K/chunk | $0.12 |
+| voyage-context-3 | Contextual document embeddings (previous gen) | 32K/chunk | $0.18 |
 
 ## Rerank
 

litellm/llms/voyage/embedding/transformation_contextual.py

@@ -105,8 +105,15 @@ def transform_embedding_request(
         optional_params: dict,
         headers: dict,
     ) -> dict:
+        # Voyage contextualized embeddings API accepts `inputs` as a list[str]
+        # (a single document's chunks). Pass list[list[str]] through unchanged
+        # only when the caller already provided that nested shape.
+        if isinstance(input, str):
+            inputs: Union[List[str], List[List[str]]] = [input]
+        else:
+            inputs = input
         return {
-            "inputs": input,
+            "inputs": inputs,
             "model": model,
             **optional_params,
         }

litellm/model_prices_and_context_window_backup.json

@@ -27345,6 +27345,38 @@
         "mode": "embedding",
         "output_cost_per_token": 0.0
     },
+    "voyage/voyage-4": {
+        "input_cost_per_token": 6e-08,
+        "litellm_provider": "voyage",
+        "max_input_tokens": 32000,
+        "max_tokens": 32000,
+        "mode": "embedding",
+        "output_cost_per_token": 0.0
+    },
+    "voyage/voyage-4-large": {
+        "input_cost_per_token": 1.2e-07,
+        "litellm_provider": "voyage",
+        "max_input_tokens": 32000,
+        "max_tokens": 32000,
+        "mode": "embedding",
+        "output_cost_per_token": 0.0
+    },
+    "voyage/voyage-4-lite": {
+        "input_cost_per_token": 2e-08,
+        "litellm_provider": "voyage",
+        "max_input_tokens": 32000,
+        "max_tokens": 32000,
+        "mode": "embedding",
+        "output_cost_per_token": 0.0
+    },
+    "voyage/voyage-4-nano": {
+        "input_cost_per_token": 0.0,
+        "litellm_provider": "voyage",
+        "max_input_tokens": 32000,
+        "max_tokens": 32000,
+        "mode": "embedding",
+        "output_cost_per_token": 0.0
+    },
     "voyage/voyage-code-2": {
         "input_cost_per_token": 1.2e-07,
         "litellm_provider": "voyage",
@@ -27369,6 +27401,14 @@
         "mode": "embedding",
         "output_cost_per_token": 0.0
     },
+    "voyage/voyage-context-4": {
+        "input_cost_per_token": 1.2e-07,
+        "litellm_provider": "voyage",
+        "max_input_tokens": 120000,
+        "max_tokens": 120000,
+        "mode": "embedding",
+        "output_cost_per_token": 0.0
+    },
     "voyage/voyage-finance-2": {
         "input_cost_per_token": 1.2e-07,
         "litellm_provider": "voyage",
@@ -27385,6 +27425,14 @@
         "mode": "embedding",
         "output_cost_per_token": 0.0
     },
+    "voyage/voyage-large-2-instruct": {
+        "input_cost_per_token": 1.2e-07,
+        "litellm_provider": "voyage",
+        "max_input_tokens": 16000,
+        "max_tokens": 16000,
+        "mode": "embedding",
+        "output_cost_per_token": 0.0
+    },
     "voyage/voyage-law-2": {
         "input_cost_per_token": 1.2e-07,
         "litellm_provider": "voyage",
@@ -27417,6 +27465,22 @@
         "mode": "embedding",
         "output_cost_per_token": 0.0
     },
+    "voyage/voyage-multimodal-3.5": {
+        "input_cost_per_token": 1.2e-07,
+        "litellm_provider": "voyage",
+        "max_input_tokens": 32000,
+        "max_tokens": 32000,
+        "mode": "embedding",
+        "output_cost_per_token": 0.0
+    },
+    "voyage/voyage-multilingual-2": {
+        "input_cost_per_token": 1.2e-07,
+        "litellm_provider": "voyage",
+        "max_input_tokens": 32000,
+        "max_tokens": 32000,
+        "mode": "embedding",
+        "output_cost_per_token": 0.0
+    },
     "wandb/openai/gpt-oss-120b": {
         "max_tokens": 131072,
         "max_input_tokens": 131072,

model_prices_and_context_window.json

@@ -27345,6 +27345,38 @@
         "mode": "embedding",
         "output_cost_per_token": 0.0
     },
+    "voyage/voyage-4": {
+        "input_cost_per_token": 6e-08,
+        "litellm_provider": "voyage",
+        "max_input_tokens": 32000,
+        "max_tokens": 32000,
+        "mode": "embedding",
+        "output_cost_per_token": 0.0
+    },
+    "voyage/voyage-4-large": {
+        "input_cost_per_token": 1.2e-07,
+        "litellm_provider": "voyage",
+        "max_input_tokens": 32000,
+        "max_tokens": 32000,
+        "mode": "embedding",
+        "output_cost_per_token": 0.0
+    },
+    "voyage/voyage-4-lite": {
+        "input_cost_per_token": 2e-08,
+        "litellm_provider": "voyage",
+        "max_input_tokens": 32000,
+        "max_tokens": 32000,
+        "mode": "embedding",
+        "output_cost_per_token": 0.0
+    },
+    "voyage/voyage-4-nano": {
+        "input_cost_per_token": 0.0,
+        "litellm_provider": "voyage",
+        "max_input_tokens": 32000,
+        "max_tokens": 32000,
+        "mode": "embedding",
+        "output_cost_per_token": 0.0
+    },
     "voyage/voyage-code-2": {
         "input_cost_per_token": 1.2e-07,
         "litellm_provider": "voyage",
@@ -27369,6 +27401,14 @@
         "mode": "embedding",
         "output_cost_per_token": 0.0
     },
+    "voyage/voyage-context-4": {
+        "input_cost_per_token": 1.2e-07,
+        "litellm_provider": "voyage",
+        "max_input_tokens": 120000,
+        "max_tokens": 120000,
+        "mode": "embedding",
+        "output_cost_per_token": 0.0
+    },
     "voyage/voyage-finance-2": {
         "input_cost_per_token": 1.2e-07,
         "litellm_provider": "voyage",
@@ -27385,6 +27425,14 @@
         "mode": "embedding",
         "output_cost_per_token": 0.0
     },
+    "voyage/voyage-large-2-instruct": {
+        "input_cost_per_token": 1.2e-07,
+        "litellm_provider": "voyage",
+        "max_input_tokens": 16000,
+        "max_tokens": 16000,
+        "mode": "embedding",
+        "output_cost_per_token": 0.0
+    },
     "voyage/voyage-law-2": {
         "input_cost_per_token": 1.2e-07,
         "litellm_provider": "voyage",
@@ -27417,6 +27465,22 @@
         "mode": "embedding",
         "output_cost_per_token": 0.0
     },
+    "voyage/voyage-multimodal-3.5": {
+        "input_cost_per_token": 1.2e-07,
+        "litellm_provider": "voyage",
+        "max_input_tokens": 32000,
+        "max_tokens": 32000,
+        "mode": "embedding",
+        "output_cost_per_token": 0.0
+    },
+    "voyage/voyage-multilingual-2": {
+        "input_cost_per_token": 1.2e-07,
+        "litellm_provider": "voyage",
+        "max_input_tokens": 32000,
+        "max_tokens": 32000,
+        "mode": "embedding",
+        "output_cost_per_token": 0.0
+    },
     "wandb/openai/gpt-oss-120b": {
         "max_tokens": 131072,
         "max_input_tokens": 131072,

tests/llm_translation/test_voyage_ai.py

@@ -142,6 +142,7 @@ def test_contextual_embedding_model_detection(self):
 
         # Test contextual model detection
         assert config.is_contextualized_embeddings("voyage-context-3") is True
+        assert config.is_contextualized_embeddings("voyage-context-4") is True
         assert config.is_contextualized_embeddings("voyage-context-2") is True
         assert config.is_contextualized_embeddings("context-model") is True
 
@@ -198,6 +199,33 @@ def test_contextual_embedding_request_transformation(self):
         assert transformed["model"] == "voyage-context-3"
         assert transformed["encoding_format"] == "float"
 
+    def test_contextual_embedding_request_input_normalization(self):
+        """Contextual API is called with list[str]; list[list[str]] only when received."""
+        from litellm.llms.voyage.embedding.transformation_contextual import (
+            VoyageContextualEmbeddingConfig,
+        )
+
+        config = VoyageContextualEmbeddingConfig()
+
+        # A bare string is normalized to a single-element list[str]
+        transformed = config.transform_embedding_request(
+            "voyage-context-4", "Hello", {}, {}
+        )
+        assert transformed["inputs"] == ["Hello"]
+
+        # A list[str] is passed through unchanged (not wrapped)
+        transformed = config.transform_embedding_request(
+            "voyage-context-4", ["Hello", "world"], {}, {}
+        )
+        assert transformed["inputs"] == ["Hello", "world"]
+
+        # A list[list[str]] is passed through unchanged
+        nested = [["Hello", "world"], ["Test"]]
+        transformed = config.transform_embedding_request(
+            "voyage-context-4", nested, {}, {}
+        )
+        assert transformed["inputs"] == nested
+
     def test_contextual_embedding_response_transformation(self):
         """Test response transformation for contextual embeddings"""
         from litellm.llms.voyage.embedding.transformation_contextual import (