# Huggingface/text-embeddings-inference, cpu bug

**URL:** <https://discuss.huggingface.co/t/huggingface-text-embeddings-inference-cpu-bug/177129>\
**Category:** Intermediate\
**Created:** [June 24, 2026, 6:53pm UTC](https://discuss.huggingface.co/t/huggingface-text-embeddings-inference-cpu-bug/177129 "2026-06-24T18:53:11Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![malaiwah](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/malaiwah/32/61130_2.png) [@malaiwah](https://discuss.huggingface.co/u/malaiwah)\
**Post date:** [June 24, 2026, 6:53pm UTC](https://discuss.huggingface.co/t/huggingface-text-embeddings-inference-cpu-bug/177129/1 "2026-06-24T18:53:11Z")

</div>

# Qwen3/Gemma3 candle skip attention masks for equal-length batches

I would like to draw your attention to this issue I recently posted to Github. [Qwen3/Gemma3 candle skip attention masks for equal-length batches · Issue #882 · huggingface/text-embeddings-inference · GitHub](https://github.com/huggingface/text-embeddings-inference/issues/882)

I also included a PR to fix the issue and throughly tested it on my machines.

When I am using CPU mode for Embeddings and have conccurency (yes, it is slow) this causes large accuracy issues.