pstore

Home

/

Main

/

Qwen2.5 32B quant comparison

Qwen2.5 32B quant comparison

Nov 30, 20251 min read

tags
  • model/quant
post_id
t3_1fkm5vd
post_code
spc
url
https://kweel.pythonanywhere.com/post/t3_1fkm5vd?file=topic_11.csv
comment_codes
  • kse: 5
  • ksr: 4
  • adm: 3
  • gut: 3
  • sfr: 3
  • add: 2
  • ewf: 2
  • kdr: 2
  • acc: 1
  • afu: 1
  • aoe: 1
  • apc: 1
  • br: 1
  • ci: 1
  • ksb: 1
  • ksk: 1
  • wok: 1

Similar to Llama 3 quants comparison, Benchmarking effects of quantization, Mistral-Large 35GB quantization, and exl2 quantization, but they document that they are using ollama as the backend, and Similar to Llama 3 quants comparison, Benchmarking effects of quantization, Mistral-Large 35GB quantization, and exl2 quantization, but they document that they are using ollama as the backend, and Ollama-MMLU-Pro as the evaluation tool. as the evaluation tool.


Graph View

Backlinks

  • DIY AI Posts RQ3
  • Qwen2.5 14B quant comparison

Created with Quartz v4.5.2 © 2025

  • GitHub
  • Email