publisher

ggml-org

3 builds measured across 13 families, and 3 derivatives of other people's models. Ranked by what machines proved, never by downloads.

measured — signed by the machine that ran eachhf.co/ggml-org/gemma-3-1b-it-GGUF:Q4_K_Mgemma-3/1b · Q4_K_M · ollama · 8,192 ctx · 12.6 tok/shf.co/ggml-org/gemma-3-1b-it-GGUF:Q8_0gemma-3/1b · Q8_0 · ollama · 8,192 ctx · 9.7 tok/shf.co/ggml-org/embeddinggemma-300M-GGUF:Q8_0embeddinggemma/300m · Q8_0 · ollama
derivatives — changes this publisher made to someone else's weightsDeepSeek-OCR-GGUFdeepseek-ocrMiMo-V2.6-Distill-Qwen-9B-GGUFmimo-v2-6Qianfan-OCR-GGUFqianfan-ocr
in the catalog, not yet measured — declared, unsignedhf.co/ggml-org/DeepSeek-OCR-GGUF:Q8_0deepseek-ocr/base~ggml-org · Q8_0 · ollama · derivativehf.co/ggml-org/Devstral-2-123B-Instruct-2512-GGUF:Q8_0devstral-2/123b-instruct-2512 · Q8_0 · ollama · quantizerhf.co/ggml-org/gemma-3-1b-it-GGUF:F16gemma-3/1b · F16 · ollama · quantizerhf.co/ggml-org/gemma-4-26B-A4B-it-GGUF:BF16gemma-4/26b-a4b · BF16 · ollama · quantizerhf.co/ggml-org/GLM-4.6V-GGUF:Q4_K_Mglm-4-6/v · Q4_K_M · ollama · quantizerhf.co/ggml-org/GLM-OCR-GGUF:Q8_0glm-ocr/base · Q8_0 · ollama · quantizerhf.co/ggml-org/GLM-OCR-GGUF:F16glm-ocr/base · F16 · ollama · quantizerhf.co/ggml-org/gpt-oss-120b-GGUF:MXFP4gpt-oss/120b · MXFP4 · ollama · quantizerhf.co/ggml-org/MiMo-V2.6-Distill-Qwen-9B-GGUF:Q8_0mimo-v2-6/9b~ggml-org-distill-qwen · Q8_0 · ollama · derivativeggml-org/MiMo-V2.6-Flash-MOPD-GGUF:Q2_K_Smimo-v2-6/flash-mopd · Q2_K_S · llamacpp · quantizerggml-org/MiMo-V2.6-Flash-MOPD-GGUF:MXFP4mimo-v2-6/flash-mopd · MXFP4 · llamacpp · quantizerggml-org/MiMo-V2.6-Flash-RL-GGUF:Q2_Kmimo-v2-6/flash-rl · Q2_K · llamacpp · quantizerggml-org/MiMo-V2.6-Flash-RL-GGUF:MXFP4mimo-v2-6/flash-rl · MXFP4 · llamacpp · quantizerggml-org/MiMo-V2.6-Pro-RL-GGUF:Q2_Kmimo-v2-6/pro-rl · Q2_K · llamacpp · quantizerggml-org/MiMo-V2.6-Pro-RL-GGUF:MXFP4mimo-v2-6/pro-rl · MXFP4 · llamacpp · quantizerggml-org/Muse-Spark-Open-816B-A42B-GGUF:Q8_0muse-spark/816b-a42b · Q8_0 · llamacpp · quantizerggml-org/Muse-Spark-Open-816B-A42B-GGUF:BF16muse-spark/816b-a42b · BF16 · llamacpp · quantizerhf.co/ggml-org/Qianfan-OCR-GGUF:Q8_0qianfan-ocr/base~ggml-org · Q8_0 · ollama · derivativehf.co/ggml-org/Qianfan-OCR-GGUF:F16qianfan-ocr/base~ggml-org · F16 · ollama · derivativehf.co/ggml-org/Qwen3.8-27B-GGUF:BF16qwen-3-8/27b · BF16 · ollama · quantizerhf.co/ggml-org/Step-3.5-Flash-GGUF:Q4_Kstep-3-5/flash · Q4_K · ollama · quantizer