Up to date
sentence-transformers Apache 2.0 out 5 years ago
22.7M parameters, text in, embeddings out, 512 tokens of context. 2 providers serve it from $0.005 per million input tokens.
Verified 2026-10-25 Rank #18 in Embeddings · score 100.1 method v1 Sources 3 Downloads 30d 1.4M · #89 of 619 tracked
Picked by rule from 6 tracked pieces: the lab's own post, the most shared, a field report, the latest.
The paraphrase-MiniLM-L6-v2 model converts sentences into a 384-dimensional dense vector space optimized for semantic similarity scoring.
The 1 pick above, plus these 5. Top is the radar's score: discussion, source authority and the lab's own posts, fading with age.
Live hosted endpoints, cheapest first, USD per million tokens. Ties break on output price, then provider name.
Hugging Face does not say who downloads a model or with what. These are the proxies: API providers listing it, quantized builds, fine-tunes/adapters/merges, and Spaces that load it.
Served by 2 API providers; 3 quantized builds for local use.
Top Spaces running it: Gradio-Blocks/pubmed-abstract-retriever ♥ 14 · pritamdeka/health-article-keyphrase-generator ♥ 8 · taskswithcode/semantic_similarity ♥ 8 · pritamdeka/pubmed-abstract-retriever ♥ 6
Weights only, an estimate: leave room for the KV cache. Method
About 0.1 GB at BF16: weights only, so leave room for the KV cache.
Estimate params × bytes per param × 1.2 · how we estimate
Weights, every dated article, and repo commits on one rail.
Teal is the weights (2022-03-02), grey ticks are repo commits.
25 derivatives on the Hub. Which build to run, by runtime: the most downloaded in each class.
From the Hugging Face model tree; quantized builds in teal, fine-tunes, merges and adapters in greys.
Everything else, folded: the full tables, every source.
It maps sentences and paragraphs to a 384 dimensional dense vector space for clustering or semantic search.[1]
Generated 2026-10-07 15:01 UTC by parasail-gemma-4-26b-a4b-it; each numeral points at the fact it rests on. Not a verdict.
Labels are evidence, not verdicts: measured a third-party leaderboard row for this model; author-reported a number the model author published in the card or eval files; structural a deterministic signal from the name, config or chat template, or a decision-model estimate. Each chip links to its source. All use cases.
Listed in Embeddings
| Pipeline tag | sentence-similarity |
|---|---|
| Library | sentence-transformers |
| Architectures | BertModel |
| Model type | bert |
| Params by dtype | F32: 22,713,216 · I64: 512 |
| Size band | S |
| MoE | no |
| Context length | 512 (config.json max_position_embeddings) |
| Papers | arXiv:1908.10084 |
| License | Apache 2.0 (apache-2.0) |
| Hub snapshot | 2026-10-25 |
vLLM: supported · BertModel · 7050ef881ff0835f6890a47db1859a24c65481d6SGLang: supported · BertModel · 773cb938fe9484a8f7e7e46fb1b1fe96412c59ed
vLLM registry (7050ef881ff0835f6890a47db1859a24c65481d6) · SGLang registry (773cb938fe9484a8f7e7e46fb1b1fe96412c59ed) · Matched by architecture class name against each project's registry on main; 'supported' means the architecture is registered, not that this checkpoint was tested.
Weights-only memory for paraphrase-MiniLM-L6-v2: params × bytes/param × 1.2 overhead. KV cache and activations are extra. How this is computed.
| BF16 (2 bytes/param) | 0.1 GB · Edge (<4 GB) |
|---|---|
| FP8 (1 byte/param) | 0.0 GB · Edge (<4 GB) |
| INT4 (0.5 bytes/param) | 0.0 GB · Edge (<4 GB) |
Weights only (params × bytes/param × 1.2 overhead); KV cache and activations extra.
finetune: 22quantized: 3
| Variant | Relation | Quant | Params | Band | License | Author | Likes | Downloads |
|---|---|---|---|---|---|---|---|---|
| maheen9/slo-matching-model | finetune | — | 22.7M | S | — | maheen9 | 0 | 7.1k |
| Xenova/paraphrase-MiniLM-L6-v2 | quantized | — | — | — | — | Xenova | 0 | 5.6k |
| mradermacher/paraphrase-MiniLM-L6-v2-i1-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 0 | 3.7k |
| mradermacher/paraphrase-MiniLM-L6-v2-GGUF | quantized | GGUF | — | — | apache-2.0 | mradermacher | 0 | 1.9k |
| rituraj18/quote-embedder-finetuned_ver_8 | finetune | — | 22.7M | S | — | rituraj18 | 0 | 629 |
| jtatman/paraphrase-minilm-l6-psychology | finetune | — | 22.7M | S | — | jtatman | 0 | 372 |
| KingTechnician/osmosis-joint-setfit | finetune | — | 22.7M | S | — | KingTechnician | 0 | 343 |
| rituraj18/quote-embedder-finetuned_ver_11 | finetune | — | 22.7M | S | — | rituraj18 | 0 | 319 |
| Jazielinho/fabric_model | finetune | — | 22.7M | S | — | Jazielinho | 1 | 249 |
| shivamgoel97/largefinetuneparagraph | finetune | — | 22.7M | S | — | shivamgoel97 | 0 | 288 |
| Fatini/ssf-retriever-modernbert-embed-base-paraphrase-MiniLM-L6-v2-epoch10 | finetune | — | 22.7M | S | — | Fatini | 0 | 282 |
| mann2107/BCMPIIRAB_MiniLM_HTTest | finetune | — | 22.7M | S | — | mann2107 | 0 | 231 |
| hskwon7/paraphrase-MiniLM-L6-v2-ft-for-etf-semantic-search | finetune | — | 22.7M | S | — | hskwon7 | 0 | 198 |
| Mr-FineTuner/Eval_02_Final_1 | finetune | — | 22.7M | S | — | Mr-FineTuner | 0 | 189 |
| rituraj18/quote-embedder-finetuned_ver_10 | finetune | — | 22.7M | S | — | rituraj18 | 0 | 168 |
| Jazielinho/fabric_model_1 | finetune | — | 22.7M | S | — | Jazielinho | 0 | 164 |
| Mr-FineTuner/Eval_02_Final_5 | finetune | — | 22.7M | S | — | Mr-FineTuner | 0 | 163 |
| aneforge/paraphrase-MiniLM-L6-v2 | finetune | — | 22.7M | S | apache-2.0 | aneforge | 0 | 137 |
| aasifali4813/bert-summarizer | finetune | — | 22.7M | S | — | aasifali4813 | 0 | 127 |
| tomasravel/modelo_finetuneado24 | finetune | — | 22.7M | S | — | tomasravel | 0 | 96 |
| Fatini/ssf-retriever-modernbert-embed-base-paraphrase-MiniLM-L6-v2 | finetune | — | 22.7M | S | — | Fatini | 0 | 92 |
| Bombek1/paraphrase-MiniLM-L6-v2-litert | finetune | — | — | — | apache-2.0 | Bombek1 | 0 | 73 |
| MykhailoS/llm_router | finetune | — | 22.7M | S | — | MykhailoS | 0 | 66 |
| ThomBors/NLBSE2026-python | finetune | — | 22.7M | S | — | ThomBors | 0 | 37 |
| Wispra-fr/setfit-url-classifier | finetune | — | 22.7M | S | — | Wispra-fr | 0 | 35 |
| Date | Commit | SHA |
|---|---|---|
| 2025-03-06 | Remove deprecated (SEB) evaluation results section | c9a2bfe |
| 2024-11-05 | Add exported openvino model 'openvino_model_qint8_quantized.xml' | 9a27583 |
| 2024-10-11 | Add new SentenceTransformer model with an openvino backend | f44a55b |
| 2024-10-11 | Add new SentenceTransformer model with an openvino backend | 0b74449 |
| 2024-10-11 | Add exported ONNX model 'model_qint8_avx512_vnni.onnx' | 4f4950b |
| 2024-10-11 | Add exported ONNX model 'model_qint8_avx512.onnx' | 5ad8dad |
| 2024-10-11 | Add exported ONNX model 'model_quint8_avx2.onnx' | b8c0df3 |
| 2024-10-11 | Add exported ONNX model 'model_qint8_arm64.onnx' | af603ab |
| 2024-10-11 | Add exported ONNX model 'model_O4.onnx' | 07595b0 |
| 2024-10-11 | Add exported ONNX model 'model_O3.onnx' | cbec798 |
| 2024-10-11 | Add exported ONNX model 'model_O2.onnx' | 1744e5a |
| 2024-10-11 | Add exported ONNX model 'model_O1.onnx' | a60fc70 |
| 2024-10-11 | Add new SentenceTransformer model with an onnx backend | 5efcc6d |
| 2024-03-27 | Adding `safetensors` variant of this model (#4) | 3bf4ae7 |
| 2024-02-12 | Update Sentence Transformers metadata (#7) | 77d17f9 |
| 2022-06-15 | Add TF weights | 68b97aa |
| 2021-08-30 | up | c4dfcde |
| 2021-08-05 | Update README.md | 990fc11 |
| 2021-06-22 | Add new SentenceTransformer model. | 7ee129e |
| 2021-05-19 | upload | 016e588 |
| 2021-05-19 | initial commit | f4fc442 |
Hosted endpoints observed for paraphrase-MiniLM-L6-v2, sorted by input price (USD per million tokens). Each row names its observer and date; prices change daily, so check the source before quoting. Cheapest observed input price: $0.0050/M.
| Provider | Model ref | Context | Quant | $/M in | $/M out | tok/s | TTFT | Tools | Status | Observed |
|---|---|---|---|---|---|---|---|---|---|---|
| DeepInfra deepinfra | sentence-transformers/paraphrase-MiniLM-L6-v2 | 512 | — | $0.0050 | — | — | — | — | live | 2026-10-24 |
| HF Inference hf | sentence-transformers/paraphrase-MiniLM-L6-v2 | — | — | — | — | — | — | — | live | 2026-10-25 |
Ranked by source signals (HN, Lobsters, authority domains, official posts) and recency. Summaries are generated from a stored copy of the article text; every sentence cites a span of that text, and nothing is shown that failed the citation check.
arXiv links found in the model card and its tags. "Peer-reviewed" is shown only with evidence: a published venue in OpenAlex (measured) or the authors' own journal reference on arXiv (reported).
Hub snapshot: 2026-10-25 · HF API: 2026-10-24 21:34 UTC · Providers: 2026-10-24 21:44 UTC · Articles: 2026-10-24 21:45 UTC · Summary: 2026-10-07 15:01 UTC · Last verified 2026-10-25
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.
