Update benchmark numbers

README.md

@@ -50,7 +50,7 @@ The `train-clean-100` variants are tuned for LibriSpeech-style English read spee
 
 ![Median latency by stage](https://huggingface.co/Retrieve0/Qwen3-ASR-1.7B-EXL3-Decoder/resolve/main/qwen3-1.7b-stage-latency.png)
 
-All configurations use all 2,620 test-clean utterances. The EXL3 decoders are about 2.0–2.1× faster than GGUF Q4_K. The official `qwen-asr` Transformers backend reaches 1.8863% WER at 863.52 ms, about 4.4× slower than the best EXL3 variant. The chart labels give WER, median latency, and the speed ratio.
+All configurations use all 2,620 test-clean utterances. The EXL3 decoders are about 1.9–2.1× faster than GGUF Q4_K. The official `qwen-asr` Transformers backend reaches 1.8863% WER at 863.52 ms, about 4.2–4.6× slower than the best EXL3 variant. The chart labels give WER, median latency, and the speed ratio.
 
 ### Requirements