Update README.md

README.md

@@ -12,6 +12,7 @@ license: apache-2.0
 short_description: All olaverse language-identification models in one Space
 models:
   - olaverse/lid-neural-5
+  - olaverse/lid-neural-5.1
   - olaverse/lid-lite-25
   - olaverse/lid-neural-25.1
   - olaverse/lid-neural-25.2
@@ -22,10 +23,11 @@ models:
 All language-identification models published under [olaverse](https://huggingface.co/olaverse), in one Space:
 
 - **[lid-neural-5](https://huggingface.co/olaverse/lid-neural-5)** — Nigerian-focused, 4 languages (Yoruba, Hausa, Igbo, Nigerian Pidgin)
+- **[lid-neural-5.1](https://huggingface.co/olaverse/lid-neural-5.1)** — same 4 languages, sentence-level tuned on `mist-encoder-base-ng` (97.6% accuracy)
 - **[lid-lite-25](https://huggingface.co/olaverse/lid-lite-25)** — fastText, CPU-only, 25 languages
 - **[lid-neural-25.1](https://huggingface.co/olaverse/lid-neural-25.1)** — XLM-R, 25 languages, tuned for passages
 - **[lid-neural-25.2](https://huggingface.co/olaverse/lid-neural-25.2)** — XLM-R, 25 languages, tuned for short queries
 
-Pick a single model to test, or run text through all four at once to compare. Only one model is kept in memory at a time to stay within hardware quota.
+Pick a single model to test, or run text through all five at once to compare. Only one model is kept in memory at a time to stay within hardware quota.
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference