Chat with ThinkingCap-Qwen3.6-27B
A finetune of Qwen 3.6 27B that keeps answer quality while cutting thinking tokens by ~50% on average (and >90% in best cases). Ask anything — reasoning, math, code, or upload an image.
Token-efficient reasoning — same answers, ~50% fewer thinking tokens
A finetune of Qwen 3.6 27B that keeps answer quality while cutting thinking tokens by ~50% on average (and >90% in best cases). Ask anything — reasoning, math, code, or upload an image.