v0.12.16

CHANGELOG.md

@@ -1,5 +1,55 @@
 # ChangeLog
 
+## [2025-01-05]
+
+### `llama-index-core` [0.12.16]
+
+- Be more lenient with leading whitespaces emitted by some models when doing ReAct (#17701)
+- Fix `user_msg` vs `chat_history` AgentWorkflow inputs (#17690)
+
+### `llama-index-embeddings-oci-data-science` [0.1.0]
+
+- Add OCI Data Science Model Deployment Embedding Integration (#17243)
+
+### `llama-index-embeddings-vllm` [0.1.0]
+
+- Add vLLM offline inference supports for embedding (#17675)
+
+### `llama-index-embeddings-voyageai` [0.3.5]
+
+- small async voyageai fix (#17698)
+
+### `llama-index-llms-gemini` [0.4.7]
+
+- gemini 2.0 support (#17720)
+- feat: support basic function call for gemini (google-generativeai) (#17696)
+
+### `llama-index-llms-oci-data-science` [0.1.0]
+
+- Add OCI Data Science Model Deployment LLM Integration (#17241)
+
+### `llama-index-llms-oci-genai` [0.3.1]
+
+- Option to pass auth_file_location, in-order to overwrite default config file location i.e. ~/.oci/config (#17695)
+
+### `llama-index-llms-ollama` [0.5.1]
+
+- fix: avoid missing tool calls while streaming
+
+### `llama-index-llms-openai` [0.3.17]
+
+- fix: max_tokens in O1 (#17703)
+- o3 mini support (#17689)
+- fix max_tokens, add reasoning_effort for openai reasoning models (#17694)
+
+### `llama-index-readers-obsidian` [0.5.0]
+
+- Improved Obsidian Reader (#17699)
+
+### `llama-index-tools-scrapegraph` [0.1.1]
+
+- feat: add new scrapegraph endpoint (#17709)
+
 ## [2025-01-31]
 
 ### `llama-index-core` [0.12.15]

docs/docs/CHANGELOG.md

@@ -1,12 +1,62 @@
 # ChangeLog
 
+## [2025-01-05]
+
+### `llama-index-core` [0.12.16]
+
+- Be more lenient with leading whitespaces emitted by some models when doing ReAct (#17701)
+- Fix `user_msg` vs `chat_history` AgentWorkflow inputs (#17690)
+
+### `llama-index-embeddings-oci-data-science` [0.1.0]
+
+- Add OCI Data Science Model Deployment Embedding Integration (#17243)
+
+### `llama-index-embeddings-vllm` [0.1.0]
+
+- Add vLLM offline inference supports for embedding (#17675)
+
+### `llama-index-embeddings-voyageai` [0.3.5]
+
+- small async voyageai fix (#17698)
+
+### `llama-index-llms-gemini` [0.4.7]
+
+- gemini 2.0 support (#17720)
+- feat: support basic function call for gemini (google-generativeai) (#17696)
+
+### `llama-index-llms-oci-data-science` [0.1.0]
+
+- Add OCI Data Science Model Deployment LLM Integration (#17241)
+
+### `llama-index-llms-oci-genai` [0.3.1]
+
+- Option to pass auth_file_location, in-order to overwrite default config file location i.e. ~/.oci/config (#17695)
+
+### `llama-index-llms-ollama` [0.5.1]
+
+- fix: avoid missing tool calls while streaming
+
+### `llama-index-llms-openai` [0.3.17]
+
+- fix: max_tokens in O1 (#17703)
+- o3 mini support (#17689)
+- fix max_tokens, add reasoning_effort for openai reasoning models (#17694)
+
+### `llama-index-readers-obsidian` [0.5.0]
+
+- Improved Obsidian Reader (#17699)
+
+### `llama-index-tools-scrapegraph` [0.1.1]
+
+- feat: add new scrapegraph endpoint (#17709)
+
 ## [2025-01-31]
 
 ### `llama-index-core` [0.12.15]
 
 - Add error_on_tool_error param to FunctionCallingLLM.predict_and_call (#17663)
 - Get tool description from pydantic field (#17679)
-- fix: make ctx._events_buffer json-serializable (#17676)
+- fix: make ctx.\_events_buffer json-serializable (#17676)
 - feat: allow to exclude empty file simple directory reader (#17656)
 - improve markdown llm output parsing (#17577)
 - small typo fix in the default plan refine prompt (#17644)

docs/docs/api_reference/embeddings/oci_data_science.md

@@ -0,0 +1,4 @@
+::: llama_index.embeddings.oci_data_science
+    options:
+      members:
+        - OCIDataScienceEmbedding

docs/docs/api_reference/embeddings/vllm.md

@@ -0,0 +1,4 @@
+::: llama_index.embeddings.vllm
+    options:
+      members:
+        - VllmEmbedding

docs/docs/api_reference/llms/oci_data_science.md

@@ -0,0 +1,4 @@
+::: llama_index.llms.oci_data_science
+    options:
+      members:
+        - OCIDataScience

docs/mkdocs.yml

@@ -261,6 +261,7 @@ nav:
           - ./examples/embeddings/nebius.ipynb
           - ./examples/embeddings/nomic.ipynb
           - ./examples/embeddings/nvidia.ipynb
+          - ./examples/embeddings/oci_data_science.ipynb
           - ./examples/embeddings/oci_genai.ipynb
           - ./examples/embeddings/ollama_embedding.ipynb
           - ./examples/embeddings/openvino.ipynb
@@ -370,6 +371,7 @@ nav:
           - ./examples/llm/nvidia_tensorrt.ipynb
           - ./examples/llm/nvidia_text_completion.ipynb
           - ./examples/llm/nvidia_triton.ipynb
+          - ./examples/llm/oci_data_science.ipynb
           - ./examples/llm/oci_genai.ipynb
           - ./examples/llm/octoai.ipynb
           - ./examples/llm/ollama.ipynb
@@ -926,6 +928,7 @@ nav:
           - ./api_reference/embeddings/nebius.md
           - ./api_reference/embeddings/nomic.md
           - ./api_reference/embeddings/nvidia.md
+          - ./api_reference/embeddings/oci_data_science.md
           - ./api_reference/embeddings/oci_genai.md
           - ./api_reference/embeddings/octoai.md
           - ./api_reference/embeddings/ollama.md
@@ -941,6 +944,7 @@ nav:
           - ./api_reference/embeddings/upstage.md
           - ./api_reference/embeddings/vertex.md
           - ./api_reference/embeddings/vertex_endpoint.md
+          - ./api_reference/embeddings/vllm.md
           - ./api_reference/embeddings/voyageai.md
           - ./api_reference/embeddings/xinference.md
           - ./api_reference/embeddings/yandexgpt.md
@@ -1044,6 +1048,7 @@ nav:
           - ./api_reference/llms/nvidia.md
           - ./api_reference/llms/nvidia_tensorrt.md
           - ./api_reference/llms/nvidia_triton.md
+          - ./api_reference/llms/oci_data_science.md
           - ./api_reference/llms/oci_genai.md
           - ./api_reference/llms/octoai.md
           - ./api_reference/llms/ollama.md
@@ -2346,6 +2351,9 @@ plugins:
             - ../llama-index-integrations/tools/llama-index-tools-linkup-research
             - ../llama-index-integrations/llms/llama-index-llms-deepseek
             - ../llama-index-integrations/llms/llama-index-llms-cortex
+            - ../llama-index-integrations/embeddings/llama-index-embeddings-vllm
+            - ../llama-index-integrations/embeddings/llama-index-embeddings-oci-data-science
+            - ../llama-index-integrations/llms/llama-index-llms-oci-data-science
   - redirects:
       redirect_maps:
         ./api/llama_index.vector_stores.MongoDBAtlasVectorSearch.html: api_reference/storage/vector_store/mongodb.md

llama-index-core/llama_index/core/__init__.py

@@ -1,6 +1,6 @@
 """Init file of LlamaIndex."""
 
-__version__ = "0.12.15"
+__version__ = "0.12.16"
 
 import logging
 from logging import NullHandler

llama-index-core/pyproject.toml

@@ -46,7 +46,7 @@ name = "llama-index-core"
 packages = [{include = "llama_index"}]
 readme = "README.md"
 repository = "https://github.com/run-llama/llama_index"
-version = "0.12.15"
+version = "0.12.16"
 
 [tool.poetry.dependencies]
 SQLAlchemy = {extras = ["asyncio"], version = ">=1.4.49"}

llama-index-integrations/embeddings/llama-index-embeddings-oci-data-science/pyproject.toml

@@ -41,9 +41,9 @@ jupyter = "^1.0.0"
 mypy = "0.991"
 pre-commit = "3.2.0"
 pylint = "2.15.10"
-pytest = "7.2.1"
+pytest = ">=7.2.1"
 pytest-asyncio = ">=0.24.0"
-pytest-mock = "3.11.1"
+pytest-mock = ">=3.11.1"
 ruff = "0.0.292"
 tree-sitter-languages = "^1.8.0"
 types-Deprecated = ">=0.1.0"

llama-index-integrations/llms/llama-index-llms-oci-data-science/pyproject.toml

@@ -39,9 +39,9 @@ jupyter = "^1.0.0"
 mypy = "0.991"
 pre-commit = "3.2.0"
 pylint = "2.15.10"
-pytest = "7.2.1"
+pytest = ">=7.2.1"
 pytest-asyncio = ">=0.24.0"
-pytest-mock = "3.11.1"
+pytest-mock = ">=3.11.1"
 ruff = "0.0.292"
 tree-sitter-languages = "^1.8.0"
 types-Deprecated = ">=0.1.0"

llama-index-integrations/llms/llama-index-llms-ollama/llama_index/llms/ollama/base.py

@@ -272,7 +272,6 @@ def get_tool_calls_from_response(
     ) -> List[ToolSelection]:
         """Predict and call the tool."""
         tool_calls = response.message.additional_kwargs.get("tool_calls", [])
-
         if len(tool_calls) < 1:
             if error_on_no_tool_call:
                 raise ValueError(
@@ -350,6 +349,8 @@ def gen() -> ChatResponseGen:
             )
 
             response_txt = ""
+            seen_tool_calls = set()
+            all_tool_calls = []
 
             for r in response:
                 if r["message"]["content"] is None:
@@ -359,7 +360,20 @@ def gen() -> ChatResponseGen:
 
                 response_txt += r["message"]["content"]
 
-                tool_calls = r["message"].get("tool_calls", [])
+                new_tool_calls = [dict(t) for t in r["message"].get("tool_calls", [])]
+                for tool_call in new_tool_calls:
+                    if (
+                        str(tool_call["function"]["name"]),
+                        str(tool_call["function"]["arguments"]),
+                    ) in seen_tool_calls:
+                        continue
+                    seen_tool_calls.add(
+                        (
+                            str(tool_call["function"]["name"]),
+                            str(tool_call["function"]["arguments"]),
+                        )
+                    )
+                    all_tool_calls.append(tool_call)
                 token_counts = self._get_response_token_counts(r)
                 if token_counts:
                     r["usage"] = token_counts
@@ -368,7 +382,7 @@ def gen() -> ChatResponseGen:
                     message=ChatMessage(
                         content=response_txt,
                         role=r["message"]["role"],
-                        additional_kwargs={"tool_calls": tool_calls},
+                        additional_kwargs={"tool_calls": list(set(all_tool_calls))},
                     ),
                     delta=r["message"]["content"],
                     raw=r,
@@ -397,6 +411,8 @@ async def gen() -> ChatResponseAsyncGen:
             )
 
             response_txt = ""
+            seen_tool_calls = set()
+            all_tool_calls = []
 
             async for r in response:
                 if r["message"]["content"] is None:
@@ -406,7 +422,20 @@ async def gen() -> ChatResponseAsyncGen:
 
                 response_txt += r["message"]["content"]
 
-                tool_calls = r["message"].get("tool_calls", [])
+                new_tool_calls = [dict(t) for t in r["message"].get("tool_calls", [])]
+                for tool_call in new_tool_calls:
+                    if (
+                        str(tool_call["function"]["name"]),
+                        str(tool_call["function"]["arguments"]),
+                    ) in seen_tool_calls:
+                        continue
+                    seen_tool_calls.add(
+                        (
+                            str(tool_call["function"]["name"]),
+                            str(tool_call["function"]["arguments"]),
+                        )
+                    )
+                    all_tool_calls.append(tool_call)
                 token_counts = self._get_response_token_counts(r)
                 if token_counts:
                     r["usage"] = token_counts
@@ -415,7 +444,7 @@ async def gen() -> ChatResponseAsyncGen:
                     message=ChatMessage(
                         content=response_txt,
                         role=r["message"]["role"],
-                        additional_kwargs={"tool_calls": tool_calls},
+                        additional_kwargs={"tool_calls": all_tool_calls},
                     ),
                     delta=r["message"]["content"],
                     raw=r,

llama-index-integrations/llms/llama-index-llms-ollama/pyproject.toml

@@ -27,7 +27,7 @@ exclude = ["**/BUILD"]
 license = "MIT"
 name = "llama-index-llms-ollama"
 readme = "README.md"
-version = "0.5.0"
+version = "0.5.1"
 
 [tool.poetry.dependencies]
 python = ">=3.9,<4.0"