LateOn Code
Coming soonEncode search queries and source code as token vectors for MaxSim retrieval.
- Released
- Feb 12, 2026
One input, different models. Compare outputs and measured speed side by side on your own device.
Model comparisons are coming with the upgraded LocalAI Tools. Explore the models below, or join the waitlist to hear when it’s ready.
Encode search queries and source code as token vectors for MaxSim retrieval.
Encode queries and passages as token vectors for MaxSim retrieval.
Convert sentences into vectors for meaning-based search, comparison, and clustering. Choose the English or multilingual variant; this model does not write answers.
Convert text into vectors for semantic search and similarity matching. Choose English or Chinese and use the appropriate query formatting.
Find text with a similar meaning across languages for search, clustering, and document matching. Converts each input into a 1024-number dense vector; this worker does not return sparse or multi-vector embeddings.
Convert multilingual text into vectors for semantic search and similarity matching. Returns normalized 768-number vectors, optionally shortened to 512, 256, or 128 values; queries and documents must use the expected formatting.
Convert multilingual text into vectors for meaning-based retrieval and comparison. Use the vectors to find similar passages; this model does not generate replies.
Represent multilingual queries, passages, documents, or code as 384- or 768-number vectors. Use these vectors for search and similarity comparisons; the model does not answer questions or execute code.
Convert multilingual text into vectors for semantic search and document matching. Choose 0.6B for a smaller download or 4B for the larger checkpoint; query instructions affect retrieval quality.