llm-gemini 0.7__tar.gz → 0.9__tar.gz

{llm_gemini-0.7 → llm_gemini-0.9}/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.2
 Name: llm-gemini
-Version: 0.7
+Version: 0.9
 Summary: LLM plugin to access Google's Gemini family of models
 Author: Simon Willison
 License: Apache-2.0
@@ -64,6 +64,9 @@ Other models are:
 - `gemini-exp-1121` - recent experimental #2
 - `gemini-exp-1206` - recent experimental #3
 - `gemini-2.0-flash-exp` - [Gemini 2.0 Flash](https://blog.google/technology/google-deepmind/google-gemini-ai-update-december-2024/#gemini-2-0-flash)
+- `learnlm-1.5-pro-experimental` - "an experimental task-specific model that has been trained to align with learning science principles" - [more details here](https://ai.google.dev/gemini-api/docs/learnlm).
+- `gemini-2.0-flash-thinking-exp-1219` - experimental "thinking" model from December 2024
+- `gemini-2.0-flash-thinking-exp-01-21` - experimental "thinking" model from January 2025
 
 ### Images, audio and video
 
@@ -113,6 +116,20 @@ To enable this feature, use `-o code_execution 1`:
 llm -m gemini-1.5-pro-latest -o code_execution 1 \
 'use python to calculate (factorial of 13) * 3'
 ```
+### Google search
+
+Some Gemini models support [Grounding with Google Search](https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/ground-gemini#web-ground-gemini), where the model can run a Google search and use the results as part of answering a prompt.
+
+Using this feature may incur additional requirements in terms of how you use the results. Consult [Google's documentation](https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/ground-gemini#web-ground-gemini) for more details.
+
+To run a prompt with Google search enabled, use `-o google_search 1`:
+
+```bash
+llm -m gemini-1.5-pro-latest -o google_search 1 \
+  'What happened in Ireland today?'
+```
+
+Use `llm logs -c --json` after running a prompt to see the full JSON response, which includes [additional information](https://github.com/simonw/llm-gemini/pull/29#issuecomment-2606201877) about grounded results.
 
 ### Chat
 

{llm_gemini-0.7 → llm_gemini-0.9}/README.md

@@ -44,6 +44,9 @@ Other models are:
 - `gemini-exp-1121` - recent experimental #2
 - `gemini-exp-1206` - recent experimental #3
 - `gemini-2.0-flash-exp` - [Gemini 2.0 Flash](https://blog.google/technology/google-deepmind/google-gemini-ai-update-december-2024/#gemini-2-0-flash)
+- `learnlm-1.5-pro-experimental` - "an experimental task-specific model that has been trained to align with learning science principles" - [more details here](https://ai.google.dev/gemini-api/docs/learnlm).
+- `gemini-2.0-flash-thinking-exp-1219` - experimental "thinking" model from December 2024
+- `gemini-2.0-flash-thinking-exp-01-21` - experimental "thinking" model from January 2025
 
 ### Images, audio and video
 
@@ -93,6 +96,20 @@ To enable this feature, use `-o code_execution 1`:
 llm -m gemini-1.5-pro-latest -o code_execution 1 \
 'use python to calculate (factorial of 13) * 3'
 ```
+### Google search
+
+Some Gemini models support [Grounding with Google Search](https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/ground-gemini#web-ground-gemini), where the model can run a Google search and use the results as part of answering a prompt.
+
+Using this feature may incur additional requirements in terms of how you use the results. Consult [Google's documentation](https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/ground-gemini#web-ground-gemini) for more details.
+
+To run a prompt with Google search enabled, use `-o google_search 1`:
+
+```bash
+llm -m gemini-1.5-pro-latest -o google_search 1 \
+  'What happened in Ireland today?'
+```
+
+Use `llm logs -c --json` after running a prompt to see the full JSON response, which includes [additional information](https://github.com/simonw/llm-gemini/pull/29#issuecomment-2606201877) about grounded results.
 
 ### Chat
 

{llm_gemini-0.7 → llm_gemini-0.9}/llm_gemini.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
-Metadata-Version: 2.1
+Metadata-Version: 2.2
 Name: llm-gemini
-Version: 0.7
+Version: 0.9
 Summary: LLM plugin to access Google's Gemini family of models
 Author: Simon Willison
 License: Apache-2.0
@@ -64,6 +64,9 @@ Other models are:
 - `gemini-exp-1121` - recent experimental #2
 - `gemini-exp-1206` - recent experimental #3
 - `gemini-2.0-flash-exp` - [Gemini 2.0 Flash](https://blog.google/technology/google-deepmind/google-gemini-ai-update-december-2024/#gemini-2-0-flash)
+- `learnlm-1.5-pro-experimental` - "an experimental task-specific model that has been trained to align with learning science principles" - [more details here](https://ai.google.dev/gemini-api/docs/learnlm).
+- `gemini-2.0-flash-thinking-exp-1219` - experimental "thinking" model from December 2024
+- `gemini-2.0-flash-thinking-exp-01-21` - experimental "thinking" model from January 2025
 
 ### Images, audio and video
 
@@ -113,6 +116,20 @@ To enable this feature, use `-o code_execution 1`:
 llm -m gemini-1.5-pro-latest -o code_execution 1 \
 'use python to calculate (factorial of 13) * 3'
 ```
+### Google search
+
+Some Gemini models support [Grounding with Google Search](https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/ground-gemini#web-ground-gemini), where the model can run a Google search and use the results as part of answering a prompt.
+
+Using this feature may incur additional requirements in terms of how you use the results. Consult [Google's documentation](https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/ground-gemini#web-ground-gemini) for more details.
+
+To run a prompt with Google search enabled, use `-o google_search 1`:
+
+```bash
+llm -m gemini-1.5-pro-latest -o google_search 1 \
+  'What happened in Ireland today?'
+```
+
+Use `llm logs -c --json` after running a prompt to see the full JSON response, which includes [additional information](https://github.com/simonw/llm-gemini/pull/29#issuecomment-2606201877) about grounded results.
 
 ### Chat
 

{llm_gemini-0.7 → llm_gemini-0.9}/llm_gemini.py

@@ -23,6 +23,17 @@ SAFETY_SETTINGS = [
     },
 ]
 
+# https://cloud.google.com/vertex-ai/generative-ai/docs/multimodal/ground-gemini#supported_models_2
+GOOGLE_SEARCH_MODELS = {
+    "gemini-1.5-pro-latest",
+    "gemini-1.5-flash-latest",
+    "gemini-1.5-pro-001",
+    "gemini-1.5-flash-001",
+    "gemini-1.5-pro-002",
+    "gemini-1.5-flash-002",
+    "gemini-2.0-flash-exp",
+}
+
 
 @llm.hookimpl
 def register_models(register):
@@ -41,8 +52,15 @@ def register_models(register):
         "gemini-exp-1121",
         "gemini-exp-1206",
         "gemini-2.0-flash-exp",
+        "learnlm-1.5-pro-experimental",
+        "gemini-2.0-flash-thinking-exp-1219",
+        "gemini-2.0-flash-thinking-exp-01-21",
     ]:
-        register(GeminiPro(model_id), AsyncGeminiPro(model_id))
+        can_google_search = model_id in GOOGLE_SEARCH_MODELS
+        register(
+            GeminiPro(model_id, can_google_search=can_google_search),
+            AsyncGeminiPro(model_id, can_google_search=can_google_search),
+        )
 
 
 def resolve_type(attachment):
@@ -50,6 +68,8 @@ def resolve_type(attachment):
     # https://github.com/simonw/llm/issues/587#issuecomment-2439785140
     if mime_type == "audio/mpeg":
         mime_type = "audio/mp3"
+    if mime_type == "application/ogg":
+        mime_type = "audio/ogg"
     return mime_type
 
 
@@ -59,6 +79,9 @@ class _SharedGemini:
     can_stream = True
 
     attachment_types = (
+        # Text
+        "text/plain",
+        "text/csv",
         # PDF
         "application/pdf",
         # Images
@@ -73,6 +96,7 @@ class _SharedGemini:
         "audio/aiff",
         "audio/aac",
         "audio/ogg",
+        "application/ogg",
         "audio/flac",
         "audio/mpeg",  # Treated as audio/mp3
         # Video
@@ -133,8 +157,17 @@ class _SharedGemini:
             default=None,
         )
 
-    def __init__(self, model_id):
+    class OptionsWithGoogleSearch(Options):
+        google_search: Optional[bool] = Field(
+            description="Enables the model to use Google Search to improve the accuracy and recency of responses from the model",
+            default=None,
+        )
+
+    def __init__(self, model_id, can_google_search=False):
         self.model_id = model_id
+        self.can_google_search = can_google_search
+        if can_google_search:
+            self.Options = self.OptionsWithGoogleSearch
 
     def build_messages(self, prompt, conversation):
         messages = []
@@ -154,7 +187,9 @@ class _SharedGemini:
                 if response.prompt.prompt:
                     parts.append({"text": response.prompt.prompt})
                 messages.append({"role": "user", "parts": parts})
-                messages.append({"role": "model", "parts": [{"text": response.text()}]})
+                messages.append(
+                    {"role": "model", "parts": [{"text": response.text_or_raise()}]}
+                )
 
         parts = []
         if prompt.prompt:
@@ -180,6 +215,8 @@ class _SharedGemini:
         }
         if prompt.options and prompt.options.code_execution:
             body["tools"] = [{"codeExecution": {}}]
+        if prompt.options and self.can_google_search and prompt.options.google_search:
+            body["tools"] = [{"google_search_retrieval": {}}]
         if prompt.system:
             body["systemInstruction"] = {"parts": [{"text": prompt.system}]}
 

{llm_gemini-0.7 → llm_gemini-0.9}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "llm-gemini"
-version = "0.7"
+version = "0.9"
 description = "LLM plugin to access Google's Gemini family of models"
 readme = "README.md"
 authors = [{name = "Simon Willison"}]