GGUF

Latest developments

Meet the Underdog Saluki 27B: A 2-bit Qwen3.8-27B That Beats the Original at Tool Calling MarkTechPost · 1 week ago · 41

Transformers now runs llama.cpp quants Hugging Face · 3 weeks ago · 20 GGUF vs GPTQ vs AWQ vs EXL2: LLM Model Formats Explained (2026) MarkTechPost · 1 month ago · 18 Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint Simon Willison’s Weblog · 1 month ago · 46 Meet S1-mini: Superwhisper’s 462 MB Open-Weights Text Normalizer That Turns Raw ASR Transcripts Into Clean Written Text MarkTechPost · 1 month ago · 51

LFM2.5 Q4\_0 Checkpoints from Quantization-Aware Distillation Hugging Face · 2 months ago · 36

Q4 2026

• 9 Oct Meet the Underdog Saluki 27B: A 2-bit Qwen3.8-27B That Beats the Original at Tool Calling

Q3 2026

◆ PrismML releases Bonsai 2 27B, an Apache 2.0 compressed multimodal LLM designed to run locally on consumer hardware Model release

• 22 Sep Transformers now runs llama.cpp quants
• 19 Sep GGUF vs GPTQ vs AWQ vs EXL2: LLM Model Formats Explained (2026)
• 17 Sep Bonsai 2 27B: Near-Lossless Compression in a 9x Smaller Footprint
• 20 Aug Meet S1-mini: Superwhisper’s 462 MB Open-Weights Text Normalizer That Turns Raw ASR Transcripts Into Clean Written Text
• 19 Aug LFM2.5 Q4\_0 Checkpoints from Quantization-Aware Distillation

Relationships

Products & technology

• Transformers integrated with this technology · 1 source
• QAD Q4_0 GGUF checkpoints integrated with this technology · 1 source
• Saluki 27B integrated with this technology · 1 source