From a1c2097aeb11fa253b2ad7586d69add31cbd4146 Mon Sep 17 00:00:00 2001 From: AnilAren <124668281+AnilAren@users.noreply.github.com> Date: Tue, 3 Jun 2025 14:42:29 +0530 Subject: [PATCH 1/2] Update supported_embedding.md --- docs/my-website/docs/embedding/supported_embedding.md | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/docs/my-website/docs/embedding/supported_embedding.md b/docs/my-website/docs/embedding/supported_embedding.md index 24c1895d1d2..789d56c9539 100644 --- a/docs/my-website/docs/embedding/supported_embedding.md +++ b/docs/my-website/docs/embedding/supported_embedding.md @@ -320,13 +320,15 @@ All models listed [here](https://build.nvidia.com/explore/retrieval) are support | NV-Embed-QA | `embedding(model="nvidia_nim/NV-Embed-QA", input)` | | nvidia/nv-embed-v1 | `embedding(model="nvidia_nim/nvidia/nv-embed-v1", input)` | | nvidia/nv-embedqa-mistral-7b-v2 | `embedding(model="nvidia_nim/nvidia/nv-embedqa-mistral-7b-v2", input)` | -| nvidia/nv-embedqa-e5-v5 | `embedding(model="nvidia_nim/nvidia/nv-embedqa-e5-v5", input)` | +| nvidia/nv-embedqa-e5-v5 | `embedding(model="nvidia_nim/nvidia/nv-embedqa-e5-v5", input, input_type="query")` | | nvidia/embed-qa-4 | `embedding(model="nvidia_nim/nvidia/embed-qa-4", input)` | | nvidia/llama-3.2-nv-embedqa-1b-v1 | `embedding(model="nvidia_nim/nvidia/llama-3.2-nv-embedqa-1b-v1", input)` | | nvidia/llama-3.2-nv-embedqa-1b-v2 | `embedding(model="nvidia_nim/nvidia/llama-3.2-nv-embedqa-1b-v2", input)` | | snowflake/arctic-embed-l | `embedding(model="nvidia_nim/snowflake/arctic-embed-l", input)` | | baai/bge-m3 | `embedding(model="nvidia_nim/baai/bge-m3", input)` | +Input type is necesary for model - "nvidia/nv-embedqa-e5-v5" + ## HuggingFace Embedding Models LiteLLM supports all Feature-Extraction + Sentence Similarity Embedding models: https://huggingface.co/models?pipeline_tag=feature-extraction From 39bb3d560c9e667a2cb6c8480b6773c93c180fed Mon Sep 17 00:00:00 2001 From: AnilAren <124668281+AnilAren@users.noreply.github.com> Date: Tue, 3 Jun 2025 15:05:23 +0530 Subject: [PATCH 2/2] Update supported_embedding.md --- .../docs/embedding/supported_embedding.md | 21 ++++++++++++++++--- 1 file changed, 18 insertions(+), 3 deletions(-) diff --git a/docs/my-website/docs/embedding/supported_embedding.md b/docs/my-website/docs/embedding/supported_embedding.md index 789d56c9539..1fd5a03e652 100644 --- a/docs/my-website/docs/embedding/supported_embedding.md +++ b/docs/my-website/docs/embedding/supported_embedding.md @@ -310,9 +310,25 @@ import os os.environ['NVIDIA_NIM_API_KEY'] = "" response = embedding( model='nvidia_nim/', - input=["good morning from litellm"] + input=["good morning from litellm"], + input_type="query" ) ``` +## `input_type` Parameter for Embedding Models + +Certain embedding models, such as `nvidia/embed-qa-4` and the E5 family, operate in **dual modes**—one for **indexing documents (passages)** and another for **querying**. To maintain high retrieval accuracy, it's essential to specify how the input text is being used by setting the `input_type` parameter correctly. + +### Usage + +Set the `input_type` parameter to one of the following values: + +- `"passage"` – for embedding content during **indexing** (e.g., documents). +- `"query"` – for embedding content during **retrieval** (e.g., user queries). + +> **Warning:** Incorrect usage of `input_type` can lead to a significant drop in retrieval performance. + + + All models listed [here](https://build.nvidia.com/explore/retrieval) are supported: | Model Name | Function Call | @@ -320,14 +336,13 @@ All models listed [here](https://build.nvidia.com/explore/retrieval) are support | NV-Embed-QA | `embedding(model="nvidia_nim/NV-Embed-QA", input)` | | nvidia/nv-embed-v1 | `embedding(model="nvidia_nim/nvidia/nv-embed-v1", input)` | | nvidia/nv-embedqa-mistral-7b-v2 | `embedding(model="nvidia_nim/nvidia/nv-embedqa-mistral-7b-v2", input)` | -| nvidia/nv-embedqa-e5-v5 | `embedding(model="nvidia_nim/nvidia/nv-embedqa-e5-v5", input, input_type="query")` | +| nvidia/nv-embedqa-e5-v5 | `embedding(model="nvidia_nim/nvidia/nv-embedqa-e5-v5", input)` | | nvidia/embed-qa-4 | `embedding(model="nvidia_nim/nvidia/embed-qa-4", input)` | | nvidia/llama-3.2-nv-embedqa-1b-v1 | `embedding(model="nvidia_nim/nvidia/llama-3.2-nv-embedqa-1b-v1", input)` | | nvidia/llama-3.2-nv-embedqa-1b-v2 | `embedding(model="nvidia_nim/nvidia/llama-3.2-nv-embedqa-1b-v2", input)` | | snowflake/arctic-embed-l | `embedding(model="nvidia_nim/snowflake/arctic-embed-l", input)` | | baai/bge-m3 | `embedding(model="nvidia_nim/baai/bge-m3", input)` | -Input type is necesary for model - "nvidia/nv-embedqa-e5-v5" ## HuggingFace Embedding Models LiteLLM supports all Feature-Extraction + Sentence Similarity Embedding models: https://huggingface.co/models?pipeline_tag=feature-extraction