kxdw2580/Qwen3-1.7B-Catgirl-test0430
TEXT GENERATIONConcurrency Cost:1Model Size:2BQuant:BF16Ctx Length:32kPublished:Apr 30, 2025License:apache-2.0Architecture:Transformer0.0K Open Weights Cold

kxdw2580/Qwen3-1.7B-Catgirl-test0430 is a 1.7 billion parameter Qwen3-based language model with a 32K context length, developed by kxdw2580. This model is a test fine-tune exploring Qwen3's capabilities compared to Qwen2.5, particularly its sensitivity to prompts and reasoning features. Preliminary evaluations indicate usable baseline performance, though it exhibits degraded long-context performance and impaired thinking-mode switching due to current fine-tuning methods.

Loading preview...