branch: elpa/gptel
commit daa2b8205183a25326d1d121dd85f4927454af0b
Author: Karthik Chikmagalur <karthikchikmaga...@gmail.com>
Commit: Karthik Chikmagalur <karthikchikmaga...@gmail.com>
gptel: Update docstring for gptel-track-media (#991)
* gptel.el (gptel-track-media): This user option is
backend-agnostic, and it should work with any model that supports
media (text, images, video etc). Update documentation to clarify
this.
---
gptel.el | 15 +++++++--------
1 file changed, 7 insertions(+), 8 deletions(-)
diff --git a/gptel.el b/gptel.el
index 7c4f8c4cd6..1f370d74d9 100644
--- a/gptel.el
+++ b/gptel.el
@@ -817,17 +817,16 @@ always handled separately."
(defcustom gptel-track-media nil
"Whether supported media in chat buffers should be sent.
-When the active `gptel-model' supports it, gptel can send images
-or other media from links in chat buffers to the LLM. To use
-this, the following steps are required.
+When the active `gptel-model' supports it, gptel can send text, images
+or other media from links in chat buffers to the LLM. To use this, the
+following steps are required.
1. `gptel-track-media' (this variable) should be non-nil
-2. The LLM should provide vision or document support. Currently,
-only the OpenAI, Anthropic and Ollama APIs are supported. See
-the documentation of `gptel-make-openai', `gptel-make-anthropic'
-and `gptel-make-ollama' resp. for details on how to specify media
-support for models.
+2. The LLM should provide vision or document support. (See
+`gptel-make-openai', `gptel-make-anthropic', `gptel-make-ollama' or
+`gptel-make-gemini' for details on how to specify media support for
+models.)
3. Only \"standalone\" links in chat buffers are considered.
These are links on their own line with no surrounding text.
