Hugging Face's logo

Spaces:
rgreenberg1
/
sparse-mpt-7b-gsm8k
Running

App Files Community
mgoin commited on
Commit
9332ccd
•
1 Parent(s): 8062816

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +3 -1
app.py CHANGED
@@ -17,8 +17,10 @@ Observe the model's performance in deciphering complex math questions, such as "
17
  ## Accelerated Inferenced on CPUs
18
  The MPT model runs purely on CPU courtesy of [software acceleration by DeepSparse](https://github.com/neuralmagic/deepsparse/tree/main/research/mpt). DeepSparse offers accelerated inference by taking advantage of the MPT model's sparsity
19
  hence delivering results fast.
20
- ![Speed](https://files.slack.com/files-pri/T020WGRLR8A-F0605DZ0B7G/image3.png?pub_secret=ab0f1d72b6)
 
21
  """
 
22
  from huggingface_hub import snapshot_download
23
  MODEL_ID = snapshot_download(repo_id=MODEL_ID, use_auth_token="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")
24
 
 
17
  ## Accelerated Inferenced on CPUs
18
  The MPT model runs purely on CPU courtesy of [software acceleration by DeepSparse](https://github.com/neuralmagic/deepsparse/tree/main/research/mpt). DeepSparse offers accelerated inference by taking advantage of the MPT model's sparsity
19
  hence delivering results fast.
20
+
21
+ ![Speedup](https://cdn-uploads.huggingface.co/production/uploads/60466e4b4f40b01b66151416/qMW-Uq8xAawhANTZYB7ZI.png)
22
  """
23
+
24
  from huggingface_hub import snapshot_download
25
  MODEL_ID = snapshot_download(repo_id=MODEL_ID, use_auth_token="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")
26