OrcaSAQ-2-27B Trends on Hugging Face: 3-Bit Mixed-Precision Qwen3-Based Reasoning Model
orcarouter · hf · 2026-09-25
OrcaSAQ-2-27B is trending on Hugging Face: a 27B text-generation model built on the Qwen3 architecture, featuring 3-bit mixed-precision quantization in safetensors format with vLLM compatibility while retaining reasoning capability. It offers an open-source option for running large reasoning models in low-VRAM environments.
More from Models
- GLiNER2.5-Decide trends on Hugging Face as open token-classification extractor — fastino · 2026-09-25
- GLiNER2.5-Decide launches: 340M classifier beats LLM routers on multi-task decisions — philipvollet · 2026-09-25
- Pure-Code Three.js Cheetah Demo Built With Opus 5.5 — majidmanzarpour · 2026-09-25
- OpenAI's 'free reset' pushed back a user's weekly reset date, unlike Anthropic's — blonderoofrat · 2026-09-25
- What's the best fully omni open model under 40B? Qwen3-Omni still leads — anovers · 2026-09-25
- Independent researcher: activation steering measures the wrong geometry — 11 experiments on Qwen2.5-7B yield AkbasCore 3.2 — Nearby_Indication474 · 2026-09-25