{"v":1,"id":"model:hf:cstr/multilingual-e5-small-GGUF","slug":"model-cstr-multilingual-e5-small-gguf","kind":"model","category":"embedding","title":"multilingual-e5-small-GGUF","summary":"GGUF format of intfloat/multilingual-e5-small for use with CrispEmbed and Ollama.","source":{"provider":"hf","ref":"cstr/multilingual-e5-small-GGUF","url":"https://huggingface.co/cstr/multilingual-e5-small-GGUF","rev":"178420da727e544c2689e89b6648b205ad176eda","fetchedAt":"2026-10-02T21:00:04.544Z","etag":"W/\"d6f-uXtVAimevsGwW6W+nAXhLqcUr1U\""},"author":{"name":"cstr","url":"https://huggingface.co/cstr"},"license":{"spdx":"mit","raw":"mit","open":true},"metrics":{"downloads":2303,"downloadsWeek":327838,"likes":1,"takenAt":"2026-10-02T21:00:04.544Z"},"tags":["gguf","embeddings","ggml","text-embeddings","bert","crispembed","ollama","feature-extraction","en","de","fr","es","zh","ja","ko","ar","hi","pt","ru","it","nl","pl","tr","vi","th","id","sv","da","no","fi","cs","ro","hu","bg","uk","ca","el","hr","sk","sl"],"pipeline":"feature-extraction","links":{"github":"ggml-org/llama.cpp","npm":"node-llama-cpp"},"updatedAt":"2026-08-04T12:35:31.000Z","collectedAt":"2026-10-02T21:00:04.544Z","review":{"numbers":["2,303 downloads on Hugging Face","1 likes","license mit","0.1 GB for multilingual-e5-small-q8_0.gguf","327,838 npm downloads a week for node-llama-cpp","latest node-llama-cpp@3.22.1"],"log":null},"trust":"unlabeled","health":{"status":"alive","checkedAt":"2026-10-02T21:00:04.544Z","http":200},"description":"# multilingual-e5-small GGUF\n\nGGUF format of [intfloat/multilingual-e5-small](https://huggingface.co/intfloat/multilingual-e5-small) for use with [CrispEmbed](https://github.com/CrispStrobe/CrispEmbed) and [Ollama](https://ollama.com).\n\n## Files\n\n| File | Quantization | Size | cos vs HF |\n|------|-------------|------|-----------|\n| [multilingual-e5-small.gguf](https://huggingface.co/cstr/multilingual-e5-small-GGUF/resolve/main/multilingual-e5-small.gguf) | F32 | 455 MiB | 1.0000 (reference) |\n| [multilingual-e5-small-q8_0.gguf](https://huggingface.co/cstr/multilingual-e5-small-GGUF/resolve/main/multilingual-e5-small-q8_0.gguf) | Q8_0 | 126 MiB | 0.9999 |\n| [multilingual-e5-small-q4_k.gguf](https://huggingface.co/cstr/multilingual-e5-small-GGUF/resolve/main/multilingual-e5-small-q4_k.gguf) | Q4_K | 115 MiB | 0.990 |\n| [multilingual-e5-small-q4_k-imatrix.gguf](https://huggingface.co/cstr/multilingual-e5-small-GGUF/resolve/main/multilingual-e5-small-q4_k-imatrix.gguf) | Q4_K + imatrix | 115 MiB | not measured |\n| [multilingual-e5-small-iq4_xs.gguf](https://huggingface.co/cstr/multilingual-e5-small-GGUF/resolve/main/multilingual-e5-small-iq4_xs.gguf) | IQ4_XS | 115 MiB | not measured |\n\nSizes are MiB (what the file browser above reports). Auxiliary files:\n[`multilingual-e5-small.imatrix`](https://huggingface.co/cstr/multilingual-e5-small-GGUF/resolve/main/multilingual-e5-small.imatrix)\n(131 KiB, the importance matrix used for the imatrix/IQ quants) and\n[`multilingual-e5-small-imatrix-ab.txt`](https://huggingface.co/cstr/multilingual-e5-small-GGUF/resolve/main/multilingual-e5-small-imatrix-ab.txt)\n(the A/B notes from producing it).\n\n**Recommended: Q8_0.** It is within 0.0001 cosine of the F32 reference while\nbeing 3.6× smaller, so F32 buys nothing you can measure. Drop to a 4-bit quant\nonly when you are memory-bound: they save a further 11 MiB — about 9% — and\nQ4_K gives up an order of magnitude more accura…\n\nSource: https://huggingface.co/cstr/multilingual-e5-small-GGUF","install":{"kind":"model","hfId":"cstr/multilingual-e5-small-GGUF","gated":false,"format":"gguf","files":[{"name":"multilingual-e5-small-iq4_xs.gguf","size":120786944,"quant":"IQ4_XS","sha256":"071b097a4de0038230005c81905d73513293d20e7dd35fd8b028a87379e06b12"},{"name":"multilingual-e5-small-q4_k-imatrix.gguf","size":121008128,"quant":"Q4_K","sha256":"0066187407d1e7df836bd1feea1c2d5d84e5d22dde22c488f25286d2500b5dbd"},{"name":"multilingual-e5-small-q4_k.gguf","size":121008128,"quant":"Q4_K","sha256":"f2f67e8536bf8cacaac14f9a66f9912121f781ba50155eae2edb74ac6e6e588f"},{"name":"multilingual-e5-small-q8_0.gguf","size":131624960,"quant":"Q8_0","sha256":"0a34067a40f25d3149b36885faa62bee0e5284d0f9edc102acfc00e115d953e8"},{"name":"multilingual-e5-small.gguf","size":476618048,"sha256":"22b4685acd49e9ba54cece93e59c1f6f721960da7ac872601ac8aef4245f31e6"}],"totalBytes":971046208,"suggestedFile":"multilingual-e5-small-q8_0.gguf","requirements":{"ramGb":1,"diskBytes":131624960,"note":"estimate: suggested file size × 1.15 + 0.5 GB; a real measurement comes with lsh models"},"runWith":["llama.cpp"],"command":"lsh models install hf:cstr/multilingual-e5-small-GGUF"}}