ukisai/Swift-Qwen3.8-27b — new model trending #30 on Hugging Face
UkisAI releases Swift-Qwen3.8-27B, a Qwen3.8-27B derivative using 58.3% fewer thinking tokens with <1% performance loss and ~1.95x speed-up.
UkisAI released Swift-Qwen3.8-27B, a reasoning-efficient derivative of Qwen3.8-27B that cuts thinking-token usage by 58.3% while staying within 1% of base performance, yielding a 1.95x speed-up on several tasks. The model was fine-tuned by penalizing reasoning-marker tokens that trigger overthinking, plus a transfer component from BottleCap AI's ThinkingCap-Qwen3.6-27B. Benchmarks include GPQA-Diamond 88.28% (base 88.38%), MMLU-Pro 84.95% (base 85.47%), and AIME 2026 94.00% (base 98.67%), with mean-token reductions of roughly 27-46% across tests. GGUF weights are available on Hugging Face alongside enterprise licensing options.
- Fine-tune reduces thinking tokens 58.3% with under 1% performance loss
- Training penalizes reasoning-marker tokens that trigger overthinking
- Includes transfer component from BottleCap AI's ThinkingCap-Qwen3.6-27B
- Mean-token reductions of 27-46% across GPQA-Diamond, MMLU-Pro, AIME, and more
From the original article
6 min read at huggingface.coimage-text-to-text · 459 downloads · 148 likes · license other
By ukisai · huggingface.co
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.