Update README.md

README.md

@@ -8645,9 +8645,11 @@ tags:
 ---
 # voyage-3-m-exp
 
-This repo contains the tokenizer of the `voyage-3-m-exp` embedding model. `voyage-3-m-exp` is an intermediate snapshot of voyage generalist embeddings optimized for academic retrieval. **For production use cases, `voyage-3-large` is highly recommended and strictly better than `voyage-3-m-exp`.** Please see the [blogpost]() for more information about `voyage-3-large`.
+This repo contains the tokenizer and evaluation results of the `voyage-3-m-exp` embedding model. `voyage-3-m-exp` is an intermediate snapshot of voyage general-purpose embeddings that are tailored to datasets similar to MTEB. The training sets of MTEB datasets are used in training `voyage-3-m-exp`.
 
-`voyage-3-m-exp` can be accessed via the [Voyage API](https://docs.voyageai.com/docs/embeddings).
+We note that for production use cases, `voyage-3-large` is highly recommended and likey strictly better than `voyage-3-m-exp`. Please see the [blogpost](https://blog.voyageai.com/2025/01/07/voyage-3-large) for more information about `voyage-3-large`.
+
+`voyage-3-m-exp` can be accessed via the [Voyage API](https://docs.voyageai.com/docs/embeddings) with the model name `"voyage-3-m-exp"`.
 
 ## Reproduction of MTEB results
 Similar to most open source models on top of the leaderboard, `voyage-3-m-exp` uses task-specific prompts. To reproduce MTEB results on the leaderboard, please set `input_type` of the API to `None` and prepend the following prompts to the input. For retrieval tasks, please use only `text` without adding the `title` field in front of the text.