add support for gemini style api, partially implements #1

README.md

@@ -44,9 +44,20 @@ See [Configuration](docs/Configuration.md).
 
 Here's an example configuration to use `gpt-4o-mini` from OpenAI. You need to set these variables in your `~/.ellrc`:
 
+```ini
+ELL_API_STYLE=gemini
+ELL_LLM_MODEL=gemini-pro
+ELL_TEMPLATE=default-gemini
+ELL_API_KEY=xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
+ELL_API_URL=https://generativelanguage.googleapis.com/v1/models/
+```
+
+Here's an example configuration to use `gpt-4o-mini` from OpenAI. 
+
 ```ini
 ELL_API_STYLE=openai
 ELL_LLM_MODEL=gpt-4o-mini
+ELL_TEMPLATE=default-openai
 ELL_API_KEY=sk-xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
 ELL_API_URL=https://api.openai.com/v1/chat/completions
 ```

ell.sh

@@ -7,7 +7,7 @@ ELL_VERSION="0.0.1";
 : "${ELL_LLM_TEMPERATURE:=0.6}";
 : "${ELL_LLM_MAX_TOKENS:=4096}";
 : "${ELL_TEMPLATE_PATH:=~/.ellrc.d/templates}";
-: "${ELL_TEMPLATE:=default}";
+: "${ELL_TEMPLATE:=default-openai}";
 : "${ELL_INPUT_FILE:=""}";
 : "${ELL_RECORD:="false"}";
 : "${ELL_INTERACTIVE:="false"}";

templates/default-gemini.json

@@ -0,0 +1,12 @@
+{
+  "contents": [
+    {
+      "role": "user",
+      "parts": [
+        {
+          "text": "You are a helpful assistant. Try to explain what happens. Follow the dialog. Be helpful. The user requests are generated with an application called \"ell\" from their command line. An example usage of this application is \"ell PROMPT\". Always ignore this command prompt and focus on other contents. Just pretend as if you never see the lines with \"ell\". Always ignore this command prompt and focus on other contents. Just pretend as if you never see the lines with \"ell\". Always ignore this command prompt and focus on other contents. Just pretend as if you never see the lines with \"ell\". Answer the user questions. If you don't know the answer, you can say \"I don't know\". \n\nContext information from terminal: \n\n---\n\n${SHELL_CONTEXT}\n\n---\n\n${USER_PROMPT}"
+        }
+      ]
+    }
+  ]
+}