model roundup

Sonnet 4.6

7 items · started 2026-08-24 · closed 2026-08-30

  1. ok bit more context: it's actually a QAT Q2 for Qwen 3.8 27 B: https://huggingface.co/sdkyuan/qwen3.8-27B-qat-q2_0-gguf QAT Q2 for DFlash model: https://huggingface.co/HermiHg/Qwen3.8-27B-DFlash2-Q2_K_S-MIX-GGUFF Q5 KV seems to cause 0 pro…

  2. I have been using Claude Code and Claude Desktop for work. In general, I prefer using these models in the chat window to understand a particular code snippet, its design, and learning new concepts.

  3. i mainly started using claude for its writing abilites and strategy when it comes to things like job outreach, important emails, cover letters, resume etc. I beleive I just used to use whatever the default model was which was Sonnet 4.6 bu…

  4. cite tags visible in the response from Claude Sonnet 5 Has anybody else been experiencing all Sonnet models show these cite tags in the past couple of days? So far Sonnet 4.6 and 5 seem to be the only models experiencing this issue.

  5. Hi all, Thought Id come here for advice from other humans. Claude seem to be reaching its limit sooner these days.

  6. So I tried Claude Pro for a month to see what Opus could do for me on the coding side. Sonnet 4.6 was my partner model that would knock out a coding prompt in maybe one or two tries, if I set the reasoning to High (or even Medium).

  7. I mostly use AI for notes explanation, PPT generation or financial models, completely non-coding. Mostly use Sonnet 5 Medium but burn credits very fast for my liking.

← all threads