cocoindex 0.1.78__tar.gz → 0.1.79__tar.gz

{cocoindex-0.1.78 → cocoindex-0.1.79}/Cargo.lock

@@ -1297,7 +1297,7 @@ dependencies = [
 
 [[package]]
 name = "cocoindex"
-version = "0.1.78"
+version = "0.1.79"
 dependencies = [
  "anyhow",
  "async-openai",

{cocoindex-0.1.78 → cocoindex-0.1.79}/Cargo.toml

@@ -2,9 +2,10 @@
 name = "cocoindex"
 # Version used for local development is always higher than others to take precedence.
 # Will be overridden for specific release versions.
-version = "0.1.78"
+version = "0.1.79"
 edition = "2024"
 rust-version = "1.88"
+readme = "README.md"
 
 [profile.release]
 codegen-units = 1

{cocoindex-0.1.78 → cocoindex-0.1.79}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cocoindex
-Version: 0.1.78
+Version: 0.1.79
 Requires-Dist: click>=8.1.8
 Requires-Dist: rich>=14.0.0
 Requires-Dist: python-dotenv>=1.1.0

{cocoindex-0.1.78 → cocoindex-0.1.79}/docs/docs/ai/llm.mdx

@@ -75,7 +75,11 @@ CocoIndex integrates with various LLM APIs for these functions.
 To use the OpenAI LLM API, you need to set the environment variable `OPENAI_API_KEY`.
 You can generate the API key from [OpenAI Dashboard](https://platform.openai.com/api-keys).
 
-Currently we don't support custom address for OpenAI API.
+If you want to use a custom address, you can either provide the `address` parameter in `LlmSpec` / `EmbedText`, or set the environment variable `OPENAI_API_BASE`. The `address` parameter takes precedence over the environment variable.
+
+Spec for OpenAI takes additional `api_config` field, in type `cocoindex.llm.OpenAiConfig` with the following fields:
+-   `org_id` (type: `str`, optional): The organization ID of the OpenAI account.
+-   `project_id` (type: `str`, optional): The project ID of the OpenAI account.
 
 You can find the full list of models supported by OpenAI [here](https://platform.openai.com/docs/models).
 

{cocoindex-0.1.78 → cocoindex-0.1.79}/docs/docs/core/settings.mdx

@@ -118,7 +118,7 @@ If you use the Postgres database hosted by [Supabase](https://supabase.com/), pl
 
 `GlobalExecutionOptions` is used to configure the global execution options shared by all flows. It has the following fields:
 
-*   `source_max_inflight_rows` (type: `int | None`, default: `256`): The maximum number of concurrent inflight rows for all source operations.
+*   `source_max_inflight_rows` (type: `int | None`, default: `1024`): The maximum number of concurrent inflight rows for all source operations.
 *   `source_max_inflight_bytes` (type: `int | None`, default: `None`): The maximum number of concurrent inflight bytes for all source operations.
 
 See also [flow definition docs](/docs/core/flow_def#control-processing-concurrency) about why it's necessary to control processing concurrency, and how to configure it on per-source basis.
@@ -136,5 +136,5 @@ This is the list of environment variables, each of which has a corresponding fie
 | `COCOINDEX_DATABASE_PASSWORD` | `database.password` | No |
 | `COCOINDEX_DATABASE_MAX_CONNECTIONS` | `database.max_connections` | No (default: `64`) |
 | `COCOINDEX_DATABASE_MIN_CONNECTIONS` | `database.min_connections` | No (default: `16`) |
-| `COCOINDEX_SOURCE_MAX_INFLIGHT_ROWS` | `global_execution_options.source_max_inflight_rows` | No (default: `256`) |
+| `COCOINDEX_SOURCE_MAX_INFLIGHT_ROWS` | `global_execution_options.source_max_inflight_rows` | No (default: `1024`) |
 | `COCOINDEX_SOURCE_MAX_INFLIGHT_BYTES` | `global_execution_options.source_max_inflight_bytes` | No |

{cocoindex-0.1.78 → cocoindex-0.1.79}/examples/image_search/colpali_main.py

@@ -27,10 +27,6 @@ COLPALI_MODEL_NAME = os.getenv("COLPALI_MODEL", "vidore/colpali-v1.2")
 print(f"📐 Using ColPali model {COLPALI_MODEL_NAME}")
 
 
-# Create ColPali embedding function using the class-based pattern
-colpali_embed = cocoindex.functions.ColPaliEmbedImage(model=COLPALI_MODEL_NAME)
-
-
 @cocoindex.transform_flow()
 def text_to_colpali_embedding(
     text: cocoindex.DataSlice[str],
@@ -56,38 +52,15 @@ def image_object_embedding_flow(
     )
     img_embeddings = data_scope.add_collector()
     with data_scope["images"].row() as img:
-        ollama_model_name = os.getenv("OLLAMA_MODEL")
-        if ollama_model_name is not None:
-            # If an Ollama model is specified, generate an image caption
-            img["caption"] = flow_builder.transform(
-                cocoindex.functions.ExtractByLlm(
-                    llm_spec=cocoindex.llm.LlmSpec(
-                        api_type=cocoindex.LlmApiType.OLLAMA, model=ollama_model_name
-                    ),
-                    instruction=(
-                        "Describe the image in one detailed sentence. "
-                        "Name all visible animal species, objects, and the main scene. "
-                        "Be specific about type, color, and notable features. "
-                        "Mention what each animal is doing."
-                    ),
-                    output_type=str,
-                ),
-                image=img["content"],
-            )
-        img["embedding"] = img["content"].transform(colpali_embed)
+        img["embedding"] = img["content"].transform(
+            cocoindex.functions.ColPaliEmbedImage(model=COLPALI_MODEL_NAME)
+        )
 
         collect_fields = {
             "id": cocoindex.GeneratedField.UUID,
             "filename": img["filename"],
             "embedding": img["embedding"],
         }
-
-        if ollama_model_name is not None:
-            print(f"Using Ollama model '{ollama_model_name}' for captioning.")
-            collect_fields["caption"] = img["caption"]
-        else:
-            print(f"No Ollama model '{ollama_model_name}' found — skipping captioning.")
-
         img_embeddings.collect(**collect_fields)
 
     img_embeddings.export(

{cocoindex-0.1.78 → cocoindex-0.1.79}/python/cocoindex/llm.py

@@ -26,6 +26,16 @@ class VertexAiConfig:
     region: str | None = None
 
 
+@dataclass
+class OpenAiConfig:
+    """A specification for a OpenAI LLM."""
+
+    kind = "OpenAi"
+
+    org_id: str | None = None
+    project_id: str | None = None
+
+
 @dataclass
 class LlmSpec:
     """A specification for a LLM."""
@@ -33,4 +43,4 @@ class LlmSpec:
     api_type: LlmApiType
     model: str
     address: str | None = None
-    api_config: VertexAiConfig | None = None
+    api_config: VertexAiConfig | OpenAiConfig | None = None

{cocoindex-0.1.78 → cocoindex-0.1.79}/python/cocoindex/setting.py

@@ -53,7 +53,7 @@ class GlobalExecutionOptions:
     """Global execution options."""
 
     # The maximum number of concurrent inflight requests, shared among all sources from all flows.
-    source_max_inflight_rows: int | None = 256
+    source_max_inflight_rows: int | None = 1024
     source_max_inflight_bytes: int | None = None
 
 

{cocoindex-0.1.78 → cocoindex-0.1.79}/src/llm/mod.rs

@@ -26,10 +26,17 @@ pub struct VertexAiConfig {
     pub region: Option<String>,
 }
 
+#[derive(Debug, Clone, Serialize, Deserialize, Default)]
+pub struct OpenAiConfig {
+    pub org_id: Option<String>,
+    pub project_id: Option<String>,
+}
+
 #[derive(Debug, Clone, Serialize, Deserialize)]
 #[serde(tag = "kind")]
 pub enum LlmApiConfig {
     VertexAi(VertexAiConfig),
+    OpenAi(OpenAiConfig),
 }
 
 #[derive(Debug, Clone, Serialize, Deserialize)]
@@ -113,7 +120,7 @@ pub async fn new_llm_generation_client(
             Box::new(ollama::Client::new(address).await?) as Box<dyn LlmGenerationClient>
         }
         LlmApiType::OpenAi => {
-            Box::new(openai::Client::new(address)?) as Box<dyn LlmGenerationClient>
+            Box::new(openai::Client::new(address, api_config)?) as Box<dyn LlmGenerationClient>
         }
         LlmApiType::Gemini => {
             Box::new(gemini::AiStudioClient::new(address)?) as Box<dyn LlmGenerationClient>
@@ -151,7 +158,7 @@ pub async fn new_llm_embedding_client(
             Box::new(gemini::AiStudioClient::new(address)?) as Box<dyn LlmEmbeddingClient>
         }
         LlmApiType::OpenAi => {
-            Box::new(openai::Client::new(address)?) as Box<dyn LlmEmbeddingClient>
+            Box::new(openai::Client::new(address, api_config)?) as Box<dyn LlmEmbeddingClient>
         }
         LlmApiType::Voyage => {
             Box::new(voyage::Client::new(address)?) as Box<dyn LlmEmbeddingClient>

{cocoindex-0.1.78 → cocoindex-0.1.79}/src/llm/openai.rs

@@ -33,17 +33,31 @@ impl Client {
         Self { client }
     }
 
-    pub fn new(address: Option<String>) -> Result<Self> {
+    pub fn new(address: Option<String>, api_config: Option<super::LlmApiConfig>) -> Result<Self> {
+        let config = match api_config {
+            Some(super::LlmApiConfig::OpenAi(config)) => config,
+            Some(_) => api_bail!("unexpected config type, expected OpenAiConfig"),
+            None => super::OpenAiConfig::default(),
+        };
+
+        let mut openai_config = OpenAIConfig::new();
         if let Some(address) = address {
-            api_bail!("OpenAI doesn't support custom API address: {address}");
+            openai_config = openai_config.with_api_base(address);
+        }
+        if let Some(org_id) = config.org_id {
+            openai_config = openai_config.with_org_id(org_id);
         }
+        if let Some(project_id) = config.project_id {
+            openai_config = openai_config.with_project_id(project_id);
+        }
+
         // Verify API key is set
         if std::env::var("OPENAI_API_KEY").is_err() {
             api_bail!("OPENAI_API_KEY environment variable must be set");
         }
         Ok(Self {
-            // OpenAI client will use OPENAI_API_KEY env variable by default
-            client: OpenAIClient::new(),
+            // OpenAI client will use OPENAI_API_KEY and OPENAI_API_BASE env variables by default
+            client: OpenAIClient::with_config(openai_config),
         })
     }
 }