̷̨̨̛̛̲͎̣̞̘͈̯͚͂̈́̂̄̽̎́̽̔͑̄̏̽̏͒̾́̅̐̈́̾̎̆͆̽́͌̽̀̚̕̚̕͠͝͝

҈̿ ̶̢̢̧̡̼̜̝̬͍̜̘̙͉̘͎͓͍̣̰͖̹͖͚̭̘̖̟͕̬̠̬͇̹̮͎̣̱̎̾͌̊́̇͛̅͂̀̀͑̃̈̀̓̏͌͌͋̐̾̒͋͋̏͋̽̈́͐͑̐̂͆̊̈́̾͌̓͌̕̚͝͝͠͠͝ͅ

Repos:

[̷̨̨̛̛̲͎̣̞̘͈̯͚͂̈́̂̄̽̎́̽̔͑̄̏̽̏͒̾́̅̐̈́̾̎̆͆̽́͌̽̀̚̕̚̕͠͝͝ @Quantum_Stat]

🚀🚀 Super excited to share the latest benchmark results for our quantized BGE models.

A few weeks ago, these models were introduced with the aim of enhancing performance and efficiency for generating embeddings. And we've now conducted thorough comparisons between running PyTorch SentenceTransformers vs. our DeepSparse-optimized models on both a 10-core laptop and a 16-core AWS instance.

The benchmarks have yielded significant improvements in processing speed. For example, running the bge-small quantized model on the 10-core laptop, achieves up to a 3X increase in speedup. What's even better, is that when tested on a 16-core AWS instance, these models achieved up to a 5X improvement.

🤗 Updated model cards:
bge-small-quant: https://t.co/UnoYEwHFNv 6K+ downloads
bge-base-quant: https://t.co/VILmxvA1pA  2K+ downloads
bge-large-quant: https://t.co/hcPrbgq05D 2K+ downloads (#1 model for STS datasets on the MTEB leaderboard)

Don't forget to check out the DeepSparse repo https://t.co/eXEZMDsBIF for more information on benchmarking and running these models on the MTEB leaderboard. 💥
cc @neuralmagic

[Raj KB @rajkbnp]

I love the #ChatGPT Cheat Sheet by Ricky Costa (@Quantum_Stat)

which includes
🔹NLP Tasks
🔹Code
🔹Structured Output Styles
🔹Unstructured Output Styles
🔹Media Types
🔹Meta ChatGPT
🔹Expert Prompting

Get your hands on this amazing resource at:https://t.co/Bg1roxcMFO

[̷̨̨̛̛̲͎̣̞̘͈̯͚͂̈́̂̄̽̎́̽̔͑̄̏̽̏͒̾́̅̐̈́̾̎̆͆̽́͌̽̀̚̕̚̕͠͝͝ @Quantum_Stat]

@DivGarg9 @MultiON_AI Like your mention of comparing it to human finger speeds, great touch!

[̷̨̨̛̛̲͎̣̞̘͈̯͚͂̈́̂̄̽̎́̽̔͑̄̏̽̏͒̾́̅̐̈́̾̎̆͆̽́͌̽̀̚̕̚̕͠͝͝ @Quantum_Stat]

@_AngelinaYang_ @MehdiAllahyari in the works, will promote when shipped :)

[̷̨̨̛̛̲͎̣̞̘͈̯͚͂̈́̂̄̽̎́̽̔͑̄̏̽̏͒̾́̅̐̈́̾̎̆͆̽́͌̽̀̚̕̚̕͠͝͝ @Quantum_Stat]

⚡IT HAPPENED!⚡

There's a new state-of-the-art sentence embeddings model for the semantic textual similarity task on Hugging Face's MTEB leaderboard 🤗!

Bge-large-en-v1.5-quant was the model I quantized in less than an hour using a single CLI command using Neural Magic's open

[̷̨̨̛̛̲͎̣̞̘͈̯͚͂̈́̂̄̽̎́̽̔͑̄̏̽̏͒̾́̅̐̈́̾̎̆͆̽́͌̽̀̚̕̚̕͠͝͝ @Quantum_Stat]

⚡Getting to Know the NPZ file format to Compress BGE Embedding Models ⚡

For One-Shot Quantization (INT8), Sparsify relies on the .npz format for data storage, a file format rooted in the mighty NumPy library.

[̷̨̨̛̛̲͎̣̞̘͈̯͚͂̈́̂̄̽̎́̽̔͑̄̏̽̏͒̾́̅̐̈́̾̎̆͆̽́͌̽̀̚̕̚̕͠͝͝ @Quantum_Stat]

Check the image below for an example of what I'm discussing 👇 We are soon releasing a notebook with an end-to-end example for anyone to replicate the compressed bge models which achieve great accuracy results on the MTEB Leaderboard.

[̷̨̨̛̛̲͎̣̞̘͈̯͚͂̈́̂̄̽̎́̽̔͑̄̏̽̏͒̾́̅̐̈́̾̎̆͆̽́͌̽̀̚̕̚̕͠͝͝ @Quantum_Stat]

GUIDE: https://t.co/Neuzi1QUTG

[̷̨̨̛̛̲͎̣̞̘͈̯͚͂̈́̂̄̽̎́̽̔͑̄̏̽̏͒̾́̅̐̈́̾̎̆͆̽́͌̽̀̚̕̚̕͠͝͝ @Quantum_Stat]

@MehdiAllahyari for compressing yes, in the works!

[̷̨̨̛̛̲͎̣̞̘͈̯͚͂̈́̂̄̽̎́̽̔͑̄̏̽̏͒̾́̅̐̈́̾̎̆͆̽́͌̽̀̚̕̚̕͠͝͝ @Quantum_Stat]

@langchain 🙃🙃🙃

[̷̨̨̛̛̲͎̣̞̘͈̯͚͂̈́̂̄̽̎́̽̔͑̄̏̽̏͒̾́̅̐̈́̾̎̆͆̽́͌̽̀̚̕̚̕͠͝͝ @Quantum_Stat]

@mrm8488 @huggingface Hey @mrm8488 , if are ever interested in training a model on financial sentiment from tweets check out my dataset, it's currently #1 twitter finance dataset: https://t.co/MF5zs40dDL

[̷̨̨̛̛̲͎̣̞̘͈̯͚͂̈́̂̄̽̎́̽̔͑̄̏̽̏͒̾́̅̐̈́̾̎̆͆̽́͌̽̀̚̕̚̕͠͝͝ @Quantum_Stat]

source library Sparsify! Not only is it ONNX and INT8 quantized (faster and lighter) but is able to run on CPUs using DeepSparse! 💥
cc @neuralmagic

Model: https://t.co/gi4exnSp0C

[̷̨̨̛̛̲͎̣̞̘͈̯͚͂̈́̂̄̽̎́̽̔͑̄̏̽̏͒̾́̅̐̈́̾̎̆͆̽́͌̽̀̚̕̚̕͠͝͝ @Quantum_Stat]

Exciting News! 🚀 DeepSparse is now integrated with @langchain , opening up a world of possibilities in Generative AI on CPUs. Langchain, known for its innovative design paradigms for large language model (LLM) applications, was often constrained by expensive APIs or cumbersome GPUs.

But with Neural Magic's DeepSparse integration, developers can now accelerate their models on CPU hardware, making it a breeze to create powerful Langchain applications.

Langchain Doc link: https://t.co/JVzAGoxsXs
DeepSparse Langchain Blog: https://t.co/6VR0bIuSfz

cc @hwchase17 @neuralmagic

[̷̨̨̛̛̲͎̣̞̘͈̯͚͂̈́̂̄̽̎́̽̔͑̄̏̽̏͒̾́̅̐̈́̾̎̆͆̽́͌̽̀̚̕̚̕͠͝͝ @Quantum_Stat]

🌟First, want to thank everyone for pushing this model past 1,000 downloads in only a few days!! Additionally, I added bge-base models to MTEB.

Most importantly, code snippets were added for running inference in the model cards for everyone to try out!

https://t.co/NZO7DPGubb

[̷̨̨̛̛̲͎̣̞̘͈̯͚͂̈́̂̄̽̎́̽̔͑̄̏̽̏͒̾́̅̐̈́̾̎̆͆̽́͌̽̀̚̕̚̕͠͝͝ @Quantum_Stat]

lol

[greg @greg16676935420]

Numbers 1-100 ranked, worst to best

100. 39
99. 41
98. 8
97. 43
96. 59
95. 74
94. 61
93. 89
92. 58
91. 12
90. 14
89. 19
88. 38
87. 71
86. 73
85. 55
84. 56
83. 37
82. 45
81. 76
80. 78
79. 96
78. 98
77. 87
76. 6
75. 68
74. 79
73. 85
72. 34
71. 42
70. 25
69. 18
68. 5
67. 84
66. 83
65. 65
64. 47
63. 53
62. 31
61. 52
60. 92
59. 91
58. 72
57. 46
56. 48
55. 23
54. 24
53. 15
52. 29
51. 51
50. 95
49. 16
48. 35
47. 64
46. 63
45. 75
44. 93
43. 13
42. 17
41. 32
40. 86
39. 62
38. 10
37. 57
36. 21
35. 20
34. 11
33. 36
32. 44
31. 90
30. 70
29. 49
28. 28
27. 27
26. 26
25. 9
24. 82
23. 81
22. 80
21. 97
20. 67
19. 66
18. 22
17. 30
16. 2
15. 50
14. 54
13. 40
12. 94
11. 60
10. 4
9. 88
8. 1
7. 77
6. 100
5. 33
4. 7
3. 99
2. 69
1. 3