全站搜索

AI 模型 84760

• zen-guard-GGUF↗

  mradermacher

• madlad400-3b-mt-Q4_K_M-GGUF↗

  mtsdurica

• Olmo-3-7B-Think-GGUF↗

  Mungert

• drex-v1.5-Q8_0↗

  nace-ai

  A Q8_0 GGUF of Drex v1.5, a decision model that takes a `state` and named typed questions and returns a probability for every option. The file is 9.5 GB, about half the size of the bf16 weights. Drex v1.5 · Drex repository · llama.cpp fork (branch `drex-v1.5`) · Ollama fork (branch `drex-v1.5`) Build the `drex-v1.5` branch of the llama.cpp fork as described on the model page. Then download the GGUF and start the server:

• OLMo-2-1124-7B-Instruct_SFT_safetyv00.01↗

  Neelectric

• gemma-4-26B-A4B-gguf↗

  Ninnix96

• d1-omni-600M-ONNX↗

  onnx-community

• HoVLE↗

  OpenGVLab

• sketch_to_image_klein_4b↗

  ostris

• pplx-embed-context-v1-0.6b↗

  perplexity-ai

  `pplx-embed-v1` and `pplx-embed-context-v1` are state-of-the-art text embedding models optimized for real-world, web-scale retrieval tasks. > [!IMPORTANT] > `pplx-embed-v1` and `pplx-embed-context-v1` natively produce *unnormalized* int8-quantized embeddings. Ensure that you compare them via *cosine similarity*. Many modern embedding models rely on instruction tuning, where users prepend an instruction string to the text being embedded. This can yield a 2%-3% lift on benchmarks, but it also introduces prompt-selection overhead and can make indexing pipelines brittle (small instruction changes can shift embedding space). We deliberately **avoid** this requirement: you can embed the text you want to index directly, without having to choose or maintain an instruction prefix.