From b93edd22f55d3e5268263c3edcdae1818505c078 Mon Sep 17 00:00:00 2001 From: Karthik Sethuraman Date: Fri, 29 Dec 2023 06:22:10 -0800 Subject: [PATCH] server : allow to generate multimodal embeddings (#4681) --- examples/server/README.md | 4 +++- examples/server/server.cpp | 12 +++++++++++- 2 files changed, 14 insertions(+), 2 deletions(-) diff --git a/examples/server/README.md b/examples/server/README.md index f1e586a1c..718a7e064 100644 --- a/examples/server/README.md +++ b/examples/server/README.md @@ -166,7 +166,7 @@ node index.js `n_probs`: If greater than 0, the response also contains the probabilities of top N tokens for each generated token (default: 0) - `image_data`: An array of objects to hold base64-encoded image `data` and its `id`s to be reference in `prompt`. You can determine the place of the image in the prompt as in the following: `USER:[img-12]Describe the image in detail.\nASSISTANT:` In this case, `[img-12]` will be replaced by the embeddings of the image id 12 in the following `image_data` array: `{..., "image_data": [{"data": "", "id": 12}]}`. Use `image_data` only with multimodal models, e.g., LLaVA. + `image_data`: An array of objects to hold base64-encoded image `data` and its `id`s to be reference in `prompt`. You can determine the place of the image in the prompt as in the following: `USER:[img-12]Describe the image in detail.\nASSISTANT:`. In this case, `[img-12]` will be replaced by the embeddings of the image with id `12` in the following `image_data` array: `{..., "image_data": [{"data": "", "id": 12}]}`. Use `image_data` only with multimodal models, e.g., LLaVA. *Result JSON:* @@ -224,6 +224,8 @@ node index.js `content`: Set the text to process. + `image_data`: An array of objects to hold base64-encoded image `data` and its `id`s to be reference in `content`. You can determine the place of the image in the content as in the following: `Image: [img-21].\nCaption: This is a picture of a house`. In this case, `[img-21]` will be replaced by the embeddings of the image with id `21` in the following `image_data` array: `{..., "image_data": [{"data": "", "id": 21}]}`. Use `image_data` only with multimodal models, e.g., LLaVA. + - **POST** `/infill`: For code infilling. Takes a prefix and a suffix and returns the predicted completion as stream. *Options:* diff --git a/examples/server/server.cpp b/examples/server/server.cpp index c5035e202..31b8cf33d 100644 --- a/examples/server/server.cpp +++ b/examples/server/server.cpp @@ -3077,7 +3077,17 @@ int main(int argc, char **argv) { prompt = ""; } - const int task_id = llama.request_completion({ {"prompt", prompt}, { "n_predict", 0} }, false, true, -1); + + json image_data; + if (body.count("image_data") != 0) { + image_data = body["image_data"]; + } + else + { + image_data = ""; + } + + const int task_id = llama.request_completion({ {"prompt", prompt}, { "n_predict", 0}, {"image_data", image_data} }, false, true, -1); task_result result = llama.next_result(task_id); return res.set_content(result.result_json.dump(), "application/json; charset=utf-8"); });