diff --git a/README.md b/README.md
index 07ca0256..0f07b73f 100644
--- a/README.md
+++ b/README.md
@@ -15,7 +15,7 @@ Its goal is to become the [AUTOMATIC1111/stable-diffusion-webui](https://github.
 * Chat mode for conversation and role-playing
 * Instruct mode compatible with various formats, including Alpaca, Vicuna, Open Assistant, Dolly, Koala, ChatGLM, MOSS, RWKV-Raven, Galactica, StableLM, WizardLM, Baize, Ziya, Chinese-Vicuna, MPT, INCITE, Wizard Mega, KoAlpaca, Vigogne, Bactrian, h2o, and OpenBuddy
 * [Multimodal pipelines, including LLaVA and MiniGPT-4](https://github.com/oobabooga/text-generation-webui/tree/main/extensions/multimodal)
-* Markdown output for [GALACTICA](https://github.com/paperswithcode/galai), including LaTeX rendering
+* Markdown output with LaTeX rendering, to use for instance with [GALACTICA](https://github.com/paperswithcode/galai)
 * Nice HTML output for GPT-4chan
 * [Custom chat characters](docs/Chat-mode.md)
 * Advanced chat features (send images, get audio responses with TTS)
diff --git a/modules/text_generation.py b/modules/text_generation.py
index d2a77ece..bc260d83 100644
--- a/modules/text_generation.py
+++ b/modules/text_generation.py
@@ -130,14 +130,11 @@ def get_reply_from_output_ids(output_ids, input_ids, original_question, state, i
 
 
 def formatted_outputs(reply, model_name):
-    if shared.model_type == 'galactica':
-        reply = fix_galactica(reply)
-        return reply, reply, generate_basic_html(reply)
-    elif shared.model_type == 'gpt4chan':
+    if shared.model_type == 'gpt4chan':
         reply = fix_gpt4chan(reply)
-        return reply, 'Only applicable for GALACTICA models.', generate_4chan_html(reply)
+        return reply, generate_4chan_html(reply)
     else:
-        return reply, 'Only applicable for GALACTICA models.', generate_basic_html(reply)
+        return reply, generate_basic_html(reply)
 
 
 def set_manual_seed(seed):
diff --git a/server.py b/server.py
index bfc72e9a..50a30268 100644
--- a/server.py
+++ b/server.py
@@ -709,6 +709,7 @@ def create_interface():
                             shared.gradio['textbox'] = gr.Textbox(value=default_text, elem_classes="textbox", lines=27)
 
                         with gr.Tab('Markdown'):
+                            shared.gradio['markdown_render'] = gr.Button('Render')
                             shared.gradio['markdown'] = gr.Markdown()
 
                         with gr.Tab('HTML'):
@@ -769,6 +770,7 @@ def create_interface():
                             shared.gradio['output_textbox'] = gr.Textbox(elem_classes="textbox_default_output", lines=27, label='Output')
 
                         with gr.Tab('Markdown'):
+                            shared.gradio['markdown_render'] = gr.Button('Render')
                             shared.gradio['markdown'] = gr.Markdown()
 
                         with gr.Tab('HTML'):
@@ -944,9 +946,9 @@ def create_interface():
         else:
             shared.input_params = [shared.gradio[k] for k in ['textbox', 'interface_state']]
             if shared.args.notebook:
-                output_params = [shared.gradio[k] for k in ['textbox', 'markdown', 'html']]
+                output_params = [shared.gradio[k] for k in ['textbox', 'html']]
             else:
-                output_params = [shared.gradio[k] for k in ['output_textbox', 'markdown', 'html']]
+                output_params = [shared.gradio[k] for k in ['output_textbox', 'html']]
 
             gen_events.append(shared.gradio['Generate'].click(
                 lambda x: x, shared.gradio['textbox'], shared.gradio['last_input']).then(
@@ -966,6 +968,7 @@ def create_interface():
 
             if shared.args.notebook:
                 shared.gradio['Undo'].click(lambda x: x, shared.gradio['last_input'], shared.gradio['textbox'], show_progress=False)
+                shared.gradio['markdown_render'].click(lambda x: x, shared.gradio['textbox'], shared.gradio['markdown'], queue=False)
                 gen_events.append(shared.gradio['Regenerate'].click(
                     lambda x: x, shared.gradio['last_input'], shared.gradio['textbox'], show_progress=False).then(
                     ui.gather_interface_values, [shared.gradio[k] for k in shared.input_elements], shared.gradio['interface_state']).then(
@@ -974,6 +977,7 @@ def create_interface():
                     # lambda: None, None, None, _js="() => {element = document.getElementsByTagName('textarea')[0]; element.scrollTop = element.scrollHeight}")
                 )
             else:
+                shared.gradio['markdown_render'].click(lambda x: x, shared.gradio['output_textbox'], shared.gradio['markdown'], queue=False)
                 gen_events.append(shared.gradio['Continue'].click(
                     ui.gather_interface_values, [shared.gradio[k] for k in shared.input_elements], shared.gradio['interface_state']).then(
                     generate_reply_wrapper, [shared.gradio['output_textbox']] + shared.input_params[1:], output_params, show_progress=False).then(