Qwen3:235b Thinking 2507 - Unsloth Dynamic 2.0 (Q3 K XL) Edit

tools thinking

Updated 3 months ago

3 months ago

0a166b05e0a4 · 103GB ·

archqwen3moe

parameters235B

quantizationQ3_K_M

103GB

Apache License Version 2.0, January 2004 http://www.apache.org/licenses/ TERMS AND CONDITIONS FOR US

11kB

{ "repeat_penalty": 1, "stop": [ "<|im_start|>", "<|im_end|>" ], "te

120B

{{- $lastUserIdx := -1 -}} {{- range $idx, $msg := .Messages -}} {{- if eq $msg.Role "user" }}{{ $la

1.7kB

Readme

Merged gguf files from Unsloth’s Thinking Q3_K_XL using the default Qwen3:255b modelfile with the recommended settings from Qwen, slightly updated by the temperature and top_p defaults recommended by Unsloth (close to the end of the page).

Find the non-thinking model here.

The performance seems to be comparable to the non-thinking model, please see the details here