llm-gemini 0.24__tar.gz → 0.26__tar.gz

{llm_gemini-0.24 → llm_gemini-0.26}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: llm-gemini
-Version: 0.24
+Version: 0.26
 Summary: LLM plugin to access Google's Gemini family of models
 Author: Simon Willison
 License-Expression: Apache-2.0
@@ -10,7 +10,7 @@ Project-URL: Issues, https://github.com/simonw/llm-gemini/issues
 Project-URL: CI, https://github.com/simonw/llm-gemini/actions
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: llm>=0.26
+Requires-Dist: llm>=0.27
 Requires-Dist: httpx
 Requires-Dist: ijson
 Provides-Extra: test
@@ -75,6 +75,9 @@ result = runner.invoke(cli.cli, ["models", "-q", "gemini/"])
 lines = reversed(result.output.strip().split("\n"))
 to_output = []
 NOTES = {
+    "gemini/gemini-flash-latest": "Latest Gemini Flash",
+    "gemini/gemini-flash-lite-latest": "Latest Gemini Flash Lite",
+    "gemini/gemini-2.5-flash": "Gemini 2.5 Flash",
     "gemini/gemini-2.5-pro": "Gemini 2.5 Pro",
     "gemini/gemini-2.5-flash": "Gemini 2.5 Flash",
     "gemini/gemini-2.5-flash-lite": "Gemini 2.5 Flash Lite",
@@ -93,6 +96,10 @@ for line in lines:
     )
 cog.out("\n".join(to_output))
 ]]] -->
+- `gemini/gemini-2.5-flash-lite-preview-09-2025`
+- `gemini/gemini-2.5-flash-preview-09-2025`
+- `gemini/gemini-flash-lite-latest`: Latest Gemini Flash Lite
+- `gemini/gemini-flash-latest`: Latest Gemini Flash
 - `gemini/gemini-2.5-flash-lite`: Gemini 2.5 Flash Lite
 - `gemini/gemini-2.5-pro`: Gemini 2.5 Pro
 - `gemini/gemini-2.5-flash`: Gemini 2.5 Flash
@@ -197,6 +204,27 @@ llm -m gemini-2.0-flash -o google_search 1 \
 
 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.
 
+### URL context
+
+Gemini models support a [URL context](https://ai.google.dev/gemini-api/docs/url-context) tool which, when enabled, allows the models to fetch additional content from URLs as part of their execution.
+
+You can enable that with the `-o url_context 1` option - for example:
+
+```bash
+llm -m gemini-2.5-flash -o url_context 1 'Latest headline on simonwillison.net'
+```
+Extra tokens introduced by this tool will be charged as input tokens. Use `--usage` to see details of those:
+```bash
+llm -m gemini-2.5-flash -o url_context 1 --usage \
+  'Latest headline on simonwillison.net'
+```
+Outputs:
+```
+The latest headline on simonwillison.net as of August 17, 2025, is "TIL: Running a gpt-oss eval suite against LM Studio on a Mac.".
+Token usage: 9,613 input, 87 output, {"candidatesTokenCount": 57, "promptTokensDetails": [{"modality": "TEXT", "tokenCount": 10}], "toolUsePromptTokenCount": 9603, "toolUsePromptTokensDetails": [{"modality": "TEXT", "tokenCount": 9603}], "thoughtsTokenCount": 30}
+```
+The `"toolUsePromptTokenCount"` key shows how many tokens were used for that URL context.
+
 ### Chat
 
 To chat interactively with the model, run `llm chat`:

{llm_gemini-0.24 → llm_gemini-0.26}/README.md

@@ -52,6 +52,9 @@ result = runner.invoke(cli.cli, ["models", "-q", "gemini/"])
 lines = reversed(result.output.strip().split("\n"))
 to_output = []
 NOTES = {
+    "gemini/gemini-flash-latest": "Latest Gemini Flash",
+    "gemini/gemini-flash-lite-latest": "Latest Gemini Flash Lite",
+    "gemini/gemini-2.5-flash": "Gemini 2.5 Flash",
     "gemini/gemini-2.5-pro": "Gemini 2.5 Pro",
     "gemini/gemini-2.5-flash": "Gemini 2.5 Flash",
     "gemini/gemini-2.5-flash-lite": "Gemini 2.5 Flash Lite",
@@ -70,6 +73,10 @@ for line in lines:
     )
 cog.out("\n".join(to_output))
 ]]] -->
+- `gemini/gemini-2.5-flash-lite-preview-09-2025`
+- `gemini/gemini-2.5-flash-preview-09-2025`
+- `gemini/gemini-flash-lite-latest`: Latest Gemini Flash Lite
+- `gemini/gemini-flash-latest`: Latest Gemini Flash
 - `gemini/gemini-2.5-flash-lite`: Gemini 2.5 Flash Lite
 - `gemini/gemini-2.5-pro`: Gemini 2.5 Pro
 - `gemini/gemini-2.5-flash`: Gemini 2.5 Flash
@@ -174,6 +181,27 @@ llm -m gemini-2.0-flash -o google_search 1 \
 
 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.
 
+### URL context
+
+Gemini models support a [URL context](https://ai.google.dev/gemini-api/docs/url-context) tool which, when enabled, allows the models to fetch additional content from URLs as part of their execution.
+
+You can enable that with the `-o url_context 1` option - for example:
+
+```bash
+llm -m gemini-2.5-flash -o url_context 1 'Latest headline on simonwillison.net'
+```
+Extra tokens introduced by this tool will be charged as input tokens. Use `--usage` to see details of those:
+```bash
+llm -m gemini-2.5-flash -o url_context 1 --usage \
+  'Latest headline on simonwillison.net'
+```
+Outputs:
+```
+The latest headline on simonwillison.net as of August 17, 2025, is "TIL: Running a gpt-oss eval suite against LM Studio on a Mac.".
+Token usage: 9,613 input, 87 output, {"candidatesTokenCount": 57, "promptTokensDetails": [{"modality": "TEXT", "tokenCount": 10}], "toolUsePromptTokenCount": 9603, "toolUsePromptTokensDetails": [{"modality": "TEXT", "tokenCount": 9603}], "thoughtsTokenCount": 30}
+```
+The `"toolUsePromptTokenCount"` key shows how many tokens were used for that URL context.
+
 ### Chat
 
 To chat interactively with the model, run `llm chat`:

{llm_gemini-0.24 → llm_gemini-0.26}/llm_gemini.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: llm-gemini
-Version: 0.24
+Version: 0.26
 Summary: LLM plugin to access Google's Gemini family of models
 Author: Simon Willison
 License-Expression: Apache-2.0
@@ -10,7 +10,7 @@ Project-URL: Issues, https://github.com/simonw/llm-gemini/issues
 Project-URL: CI, https://github.com/simonw/llm-gemini/actions
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: llm>=0.26
+Requires-Dist: llm>=0.27
 Requires-Dist: httpx
 Requires-Dist: ijson
 Provides-Extra: test
@@ -75,6 +75,9 @@ result = runner.invoke(cli.cli, ["models", "-q", "gemini/"])
 lines = reversed(result.output.strip().split("\n"))
 to_output = []
 NOTES = {
+    "gemini/gemini-flash-latest": "Latest Gemini Flash",
+    "gemini/gemini-flash-lite-latest": "Latest Gemini Flash Lite",
+    "gemini/gemini-2.5-flash": "Gemini 2.5 Flash",
     "gemini/gemini-2.5-pro": "Gemini 2.5 Pro",
     "gemini/gemini-2.5-flash": "Gemini 2.5 Flash",
     "gemini/gemini-2.5-flash-lite": "Gemini 2.5 Flash Lite",
@@ -93,6 +96,10 @@ for line in lines:
     )
 cog.out("\n".join(to_output))
 ]]] -->
+- `gemini/gemini-2.5-flash-lite-preview-09-2025`
+- `gemini/gemini-2.5-flash-preview-09-2025`
+- `gemini/gemini-flash-lite-latest`: Latest Gemini Flash Lite
+- `gemini/gemini-flash-latest`: Latest Gemini Flash
 - `gemini/gemini-2.5-flash-lite`: Gemini 2.5 Flash Lite
 - `gemini/gemini-2.5-pro`: Gemini 2.5 Pro
 - `gemini/gemini-2.5-flash`: Gemini 2.5 Flash
@@ -197,6 +204,27 @@ llm -m gemini-2.0-flash -o google_search 1 \
 
 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.
 
+### URL context
+
+Gemini models support a [URL context](https://ai.google.dev/gemini-api/docs/url-context) tool which, when enabled, allows the models to fetch additional content from URLs as part of their execution.
+
+You can enable that with the `-o url_context 1` option - for example:
+
+```bash
+llm -m gemini-2.5-flash -o url_context 1 'Latest headline on simonwillison.net'
+```
+Extra tokens introduced by this tool will be charged as input tokens. Use `--usage` to see details of those:
+```bash
+llm -m gemini-2.5-flash -o url_context 1 --usage \
+  'Latest headline on simonwillison.net'
+```
+Outputs:
+```
+The latest headline on simonwillison.net as of August 17, 2025, is "TIL: Running a gpt-oss eval suite against LM Studio on a Mac.".
+Token usage: 9,613 input, 87 output, {"candidatesTokenCount": 57, "promptTokensDetails": [{"modality": "TEXT", "tokenCount": 10}], "toolUsePromptTokenCount": 9603, "toolUsePromptTokensDetails": [{"modality": "TEXT", "tokenCount": 9603}], "thoughtsTokenCount": 30}
+```
+The `"toolUsePromptTokenCount"` key shows how many tokens were used for that URL context.
+
 ### Chat
 
 To chat interactively with the model, run `llm chat`:

{llm_gemini-0.24 → llm_gemini-0.26}/llm_gemini.egg-info/requires.txt

@@ -1,4 +1,4 @@
-llm>=0.26
+llm>=0.27
 httpx
 ijson
 

{llm_gemini-0.24 → llm_gemini-0.26}/llm_gemini.py

@@ -45,6 +45,10 @@ GOOGLE_SEARCH_MODELS = {
     "gemini-2.5-pro",
     "gemini-2.5-flash",
     "gemini-2.5-flash-lite",
+    "gemini-flash-latest",
+    "gemini-flash-lite-latest",
+    "gemini-2.5-flash-preview-09-2025",
+    "gemini-2.5-flash-lite-preview-09-2025",
 }
 
 # Older Google models used google_search_retrieval instead of google_search
@@ -70,6 +74,10 @@ THINKING_BUDGET_MODELS = {
     "gemini-2.5-pro",
     "gemini-2.5-flash",
     "gemini-2.5-flash-lite",
+    "gemini-flash-latest",
+    "gemini-flash-lite-latest",
+    "gemini-2.5-flash-preview-09-2025",
+    "gemini-2.5-flash-lite-preview-09-2025",
 }
 
 NO_VISION_MODELS = {"gemma-3-1b-it", "gemma-3n-e4b-it"}
@@ -156,6 +164,11 @@ def register_models(register):
         "gemini-2.5-pro",
         # 22nd July 2025:
         "gemini-2.5-flash-lite",
+        # 25th Spetember 2025:
+        "gemini-flash-latest",
+        "gemini-flash-lite-latest",
+        "gemini-2.5-flash-preview-09-2025",
+        "gemini-2.5-flash-lite-preview-09-2025",
     ):
         can_google_search = model_id in GOOGLE_SEARCH_MODELS
         can_thinking_budget = model_id in THINKING_BUDGET_MODELS
@@ -272,6 +285,13 @@ class _SharedGemini:
             ),
             default=None,
         )
+        url_context: Optional[bool] = Field(
+            description=(
+                "Enable the URL context tool so the model can fetch content "
+                "from URLs mentioned in the prompt"
+            ),
+            default=None,
+        )
 
     class OptionsWithGoogleSearch(Options):
         google_search: Optional[bool] = Field(
@@ -404,6 +424,8 @@ class _SharedGemini:
                 else "google_search"
             )
             tools.append({tool_name: {}})
+        if prompt.options and prompt.options.url_context:
+            tools.append({"url_context": {}})
         if prompt.tools:
             tools.append(
                 {
@@ -489,6 +511,12 @@ class _SharedGemini:
             candidates_token_count = usage.get("candidatesTokenCount") or 0
             thoughts_token_count = usage.get("thoughtsTokenCount") or 0
             output_tokens = candidates_token_count + thoughts_token_count
+            tool_token_count = usage.get("toolUsePromptTokenCount") or 0
+            if tool_token_count:
+                if input_tokens is None:
+                    input_tokens = tool_token_count
+                else:
+                    input_tokens += tool_token_count
             usage.pop("totalTokenCount", None)
             if input_tokens is not None:
                 response.set_usage(
@@ -528,6 +556,8 @@ class GeminiPro(_SharedGemini, llm.KeyModel):
                         gathered.append(event)
                     events.clear()
         response.response_json = gathered[-1]
+        resolved_model = gathered[-1]["modelVersion"]
+        response.set_resolved_model(resolved_model)
         self.set_usage(response)
 
 

{llm_gemini-0.24 → llm_gemini-0.26}/pyproject.toml

@@ -1,13 +1,13 @@
 [project]
 name = "llm-gemini"
-version = "0.24"
+version = "0.26"
 description = "LLM plugin to access Google's Gemini family of models"
 readme = "README.md"
 authors = [{name = "Simon Willison"}]
 license = "Apache-2.0"
 classifiers = []
 dependencies = [
-    "llm>=0.26",
+    "llm>=0.27",
     "httpx",
     "ijson"
 ]

{llm_gemini-0.24 → llm_gemini-0.26}/tests/test_gemini.py

@@ -242,6 +242,14 @@ def test_cli_gemini_models(tmpdir, monkeypatch):
         assert "embedContent" in model["supportedGenerationMethods"]
 
 
+@pytest.mark.vcr
+def test_resolved_model():
+    model = llm.get_model("gemini-flash-latest")
+    response = model.prompt("hi", key=GEMINI_API_KEY)
+    response.text()
+    assert response.resolved_model == "gemini-2.5-flash-preview-09-2025"
+
+
 @pytest.mark.vcr
 def test_tools():
     model = llm.get_model("gemini-2.0-flash")