20+ models
pplx-decider is a 27B decision model from Perplexity, fine-tuned from Qwen3.8-27B
Text-only Qwen3.5 2B GGUF files. Default: q5_k_m. TurinTech models, built with Artemis.
pulled as BF16, vocab slice (combined_en_huge) requant to IQ4_NL (combined_en_huge) -- (English + Coding focus) - ~20% faster than Q4_K_M
Huihui Qwen3.5 2B abliterated, GGUF Q4_K_M. A standalone release packaged by Interim Labs.
Vision removed Qwen3.5-2B.
Qwen3.5 2B in GGUF Q4_K_M, packaged by Interim Labs for local text generation.
full BF16 model, hand-requant to IQ4_NL, with combined_en_huge dataset (English + Coding focus)
Fine-tuned on top of Qwen/Qwen3.5-2B-Base with extended 5-epoch training and hard-negative mining. Optimized for structured schema-guided routing and intent classification.
Qwen3.5 2B in Q8_0 quantization. Strong balance of capability and efficiency with 262K context, vision, tool use, and thinking. Ideal for local deployment on consumer hardware.
A compact Havenlon-focused Qwen3.5 2B model for execution control, intent-to-action reasoning, policy boundaries, and AI Agent execution risk.
WSAI-Heart-Pro 是 WSAI-Heart 系列的 Pro 级正式版本，基于 Qwen3.5-2B-Base 全精度 LoRA 微调。约 2B 参数，端侧部署友好。
WSAI-Heart-Pro-Exp 是 WSAI-Heart-Pro 的实验版本，基于 Qwen3.5-2B-Base 进行全精度 LoRA 微调。相比0.8B，这次升级到了 2B 参数，共情表达与对话质量均有明显提升，且仍然可以端侧部署。
German-OCR-Turbo ist ein fine-tuned Vision-Language-Modell basierend auf Qwen3-VL-2B, optimiert für die präzise Texterkennung aus deutschen Rechnungen, Formularen und Geschäftsdokumenten. Das Modell extrahiert strukturierte Daten im Markdown-Format.
26.3.7. Update: This model introduces higher-quality reasoning trajectories across domains such as science, instruction-following, and mathematics.
Qwen3.5 2B LoRA trained to act like Rocky from Project Hail Mary by Andy Weir.
Qwen-SEA-LION-v4-32B-IT is a multilingual model which has been pretrained and instruct-tuned for the Southeast Asia region. Developed by AI Singapore and funded by National Research Foundation, Singapore.
Recommended for transcribing and summarizing text from screenshots.