huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-05-07 10:22:09 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fixing the router + template for Qwen3. (#3200)

Browse Source
This commit is contained in:
Nicolas Patry 2025-04-29 16:29:26 +02:00 committed by GitHub
parent 39cfe232fd
commit e7329fec18
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
2 changed files with 5 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
router/src/config.rs
View File

@ -414,6 +414,7 @@ pub enum Config {
T5, T5,
DeepseekV2, DeepseekV2,
DeepseekV3, DeepseekV3,
Qwen3,
} }
#[derive(Clone, Debug, Serialize, Deserialize)] #[derive(Clone, Debug, Serialize, Deserialize)]

4
router/src/infer/chat_template.rs
View File

@ -41,6 +41,10 @@ impl ChatTemplate {
"messages[0]['content'][0]['text']", "messages[0]['content'][0]['text']",
"messages[0]['content']", "messages[0]['content']",
); );
// Hack to fix Qwen3 templating.
// It uses python notation to reverse lists, which do not exist in minijinja
// so we're using the reverse filter instead.
let mutated_template = mutated_template.replace("[::-1]", "|reverse");
let template_str = mutated_template.into_boxed_str(); let template_str = mutated_template.into_boxed_str();
env.add_function("raise_exception", raise_exception); env.add_function("raise_exception", raise_exception);