infra: prepare for release 0.4.26 [generated] [skip ci]

Signed-off-by: Aaron Pham <29749331+aarnphm@users.noreply.github.com>

openllm-python/CHANGELOG.md

@@ -18,6 +18,17 @@ This changelog is managed by towncrier and is compiled at release time.
 
 <!-- towncrier release notes start -->
 
+## [0.4.26](https://github.com/bentoml/openllm/tree/v0.4.26)
+
+### Features
+
+- `/v1/chat/completions` now accepts two additional parameters
+
+  - `chat_templates`: this is a string of [Jinja templates](https://huggingface.co/docs/transformers/main/chat_templating#templates-for-chat-models)
+                       to use with this models. By default, it will just use the default models provided chat templates based on config.json.
+  - `add_generation_prompt`: See [here](https://huggingface.co/docs/transformers/main/chat_templating#how-do-i-use-chat-templates)
+  [#725](https://github.com/bentoml/openllm/issues/725)
+
 ## [0.4.25](https://github.com/bentoml/openllm/tree/v0.4.25)
 No significant changes.
 

openllm-python/pyproject.toml

@@ -40,8 +40,8 @@ classifiers = [
 dependencies = [
     "bentoml[io]>=1.1.10",
     "transformers[torch,tokenizers]>=4.35.0",
-    "openllm-client>=0.4.25",
-    "openllm-core>=0.4.25",
+    "openllm-client>=0.4.26",
+    "openllm-core>=0.4.26",
     "safetensors",
     "optimum>=1.12.0",
     "accelerate",