Hugging Face's logo

warshanks
/
Voxtral-Mini-3B-2507-FP8

Audio-Text-to-Text
vllm
Model card Files
xet
Community
warshanks commited on
Commit
a9c2856
·
verified ·
1 Parent(s): 1da1107

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -1
README.md CHANGED
@@ -18,7 +18,9 @@ pipeline_tag: audio-text-to-text
18
  base_model: mistralai/Voxtral-Mini-3B-2507
19
  ---
20
 
21
- FP8 quant of Voxtral Mini. Whisper layers were ignored and are unquantized.
 
 
22
 
23
  # Voxtral Mini 1.0 (3B) - 2507
24
 
 
18
  base_model: mistralai/Voxtral-Mini-3B-2507
19
  ---
20
 
21
+ FP8 quant of Voxtral Mini using the example script from [llm-compressor](https://github.com/vllm-project/llm-compressor/tree/main/experimental/mistral).
22
+
23
+ Whisper layers were ignored and are unquantized.
24
 
25
  # Voxtral Mini 1.0 (3B) - 2507
26