Use cases › Capability
Embed text for search
Embedding models that turn text (or images) into vectors for semantic search, RAG retrieval and clustering.
Benchmarks that represent this job 9
A model's row on one of these third-party boards is what earns the measured label; the same benchmark name in the model's own card earns author-reported.
Models 84
Narrow by what you can run and pay for. Rows are ordered measured → author-reported → structural, then by watch-list score; click a column to re-sort. Every chip links to its source.
| Model | Evidence | Params | Context | Price from | Runs on | License | Providers | Measured latency |
|---|---|---|---|---|---|---|---|---|
| bge-m3 BAAI | measured MTEB(eng, v2) partial 63.21decision-model estimate p=1.00 +1 | — | 8K | $0.01 | — | MIT | 3 | — |
| all-mpnet-base-v2 sentence-transformers | measured MTEB(eng, v2) retrieval 44.54MTEB(eng, v2) partial 59.62 +1 | 109M | 514 | $0.0050 | Laptop / edge | Apache 2.0 | 2 | — |
| paraphrase-multilingual-MiniLM-L12-v2 sentence-transformers | measured MTEB(eng, v2) retrieval 35.93MTEB(eng, v2) 56.42 +1 | 118M | 512 | — | Laptop / edge | Apache 2.0 | 1 | — |
| multilingual-e5-large intfloat | measured MTEB(eng, v2) retrieval 51.53MTEB(eng, v2) partial 62.07 +1 | 560M | 514 | $0.01 | Laptop / edge | MIT | 2 | — |
| embeddinggemma-300m google | measured MTEB(eng, v2) retrieval 55.69MTEB(eng, v2) 69.67 +1 | 303M | 2K | $0.0020 | Laptop / edge | Gemma Terms of Use attribution | 2 | — |
| Qwen3-Embedding-0.6B Qwen | measured MTEB(eng, v2) retrieval 61.83MTEB(eng, v2) 70.47 +1 | 596M | 32K | $0.01 | Laptop / edge | Apache 2.0 | 1 | — |
| bge-small-en-v1.5 BAAI | measured MTEB(eng, v2) retrieval 53.86MTEB(eng, v2) 64.3 +1 | 33.4M | 512 | — | Laptop / edge | MIT | 1 | — |
| bge-base-en-v1.5 BAAI | measured MTEB(eng, v2) retrieval 54.75MTEB(eng, v2) 65.14 +1 | 109M | 512 | $0.0050 | Laptop / edge | MIT | 2 | — |
| bge-large-en-v1.5 BAAI | measured MTEB(eng, v2) retrieval 55.44MTEB(eng, v2) 65.89 +1 | 335M | 512 | $0.01 | Laptop / edge | MIT | 2 | — |
| Qwen3-Embedding-8B Qwen | measured MTEB(eng, v2) retrieval 69.44MTEB(eng, v2) 75.22 +1 | 7.57B | 40K | $0.01 | Single GPU | Apache 2.0 | 2 | — |
| multilingual-e5-small intfloat | measured MTEB(eng, v2) retrieval 46.43MTEB(eng, v2) 59.69 +1 | 118M | 512 | — | Laptop / edge | MIT | 1 | — |
| multilingual-e5-large-instruct intfloat | measured MTEB(eng, v2) retrieval 53.47MTEB(eng, v2) 65.56 +1 | 560M | 514 | $0.01 | Laptop / edge | MIT | 2 | — |
| nomic-embed-text-v1.5 nomic-ai | measured MTEB(eng, v2) retrieval 47.97MTEB(eng, v2) 62.2 +1 | 137M | 2K | — | Laptop / edge | Apache 2.0 | 0 | — |
| paraphrase-multilingual-mpnet-base-v2 sentence-transformers | measured MTEB(eng, v2) retrieval 37.14MTEB(eng, v2) 58.27 +1 | 278M | 514 | — | Laptop / edge | Apache 2.0 | 1 | — |
| multilingual-e5-base intfloat | measured MTEB(eng, v2) retrieval 48.97MTEB(eng, v2) 61.0 +1 | 278M | 514 | — | Laptop / edge | MIT | 1 | — |
| e5-large-v2 intfloat | measured MTEB(eng, v2) retrieval 49.31MTEB(eng, v2) 62.79 +1 | 335M | 512 | $0.01 | Laptop / edge | MIT | 2 | — |
| e5-base-v2 intfloat | measured MTEB(eng, v2) retrieval 49.67MTEB(eng, v2) 62.67 +1 | 109M | 512 | $0.0050 | Laptop / edge | MIT | 2 | — |
| jina-embeddings-v3 jinaai | measured MTEB(eng, v2) retrieval 54.29MTEB(eng, v2) partial 66.6 +1 | 572M | 8K | — | Laptop / edge | CC BY-NC 4.0 non-commercial | 0 | — |
| Qwen3-Embedding-4B Qwen | measured MTEB(eng, v2) retrieval 68.46MTEB(eng, v2) 74.6 +1 | 4.02B | 40K | $0.02 | Single GPU | Apache 2.0 | 1 | — |
| nomic-embed-text-v1 nomic-ai | measured MTEB(eng, v2) retrieval 50.72MTEB(eng, v2) 63.6 +1 | 137M | 8K | — | Laptop / edge | Apache 2.0 | 0 | — |
| snowflake-arctic-embed-l-v2.0 Snowflake | measured MTEB(eng, v2) retrieval 58.56MTEB(eng, v2) 63.6 +1 | 568M | 8K | — | Laptop / edge | Apache 2.0 | 1 | — |
| nomic-embed-text-v2-moe nomic-ai | measured MTEB(eng, v2) retrieval 54.81MTEB(eng, v2) 63.64 +1 | 475M | 2K | — | Laptop / edge | Apache 2.0 | 0 | — |
| e5-small-v2 intfloat | measured MTEB(eng, v2) retrieval 48.46MTEB(eng, v2) 61.32 +1 | 33.4M | 512 | — | Laptop / edge | MIT | 1 | — |
| gte-multilingual-base Alibaba-NLP | measured MTEB(eng, v2) retrieval 50.87MTEB(eng, v2) partial 62.51 +1 | 305M | 8K | — | Laptop / edge | Apache 2.0 | 0 | — |
| embeddinggemma-2 google | measured MTEB(eng, v2) retrieval 51.33MTEB(eng, v2) 67.53 +1 | 744M | — | — | Laptop / edge | Apache 2.0 | 0 | — |
| granite-embedding-small-english-r2 ibm-granite | measured MTEB(eng, v2) retrieval 53.93MTEB(eng, v2) 61.07 +1 | 47.7M | 8K | — | Laptop / edge | Apache 2.0 | 0 | — |
| bge-base-en BAAI | measured MTEB(eng, v2) partial 72.33decision-model estimate p=1.00 +1 | 109M | 512 | — | Laptop / edge | MIT | 1 | — |
| LaBSE sentence-transformers | measured MTEB(eng, v2) retrieval 18.05MTEB(eng, v2) 48.81 +1 | 471M | 512 | — | Laptop / edge | Apache 2.0 | 1 | — |
| multi-qa-MiniLM-L6-cos-v1 sentence-transformers | measured MTEB(eng, v2) partial 67.83decision-model estimate p=1.00 +1 | 22.7M | 512 | — | Laptop / edge | Not declared review | 1 | — |
| jina-embeddings-v2-small-en jinaai | measured MTEB(eng, v2) retrieval 26.47MTEB(eng, v2) partial 51.02 +1 | 32.7M | — | — | Laptop / edge | Apache 2.0 | 0 | — |
| gte-large thenlper | measured MTEB(eng, v2) retrieval 53.29MTEB(eng, v2) 64.77 +1 | 335M | 512 | $0.01 | Laptop / edge | MIT | 2 | — |
| bge-multilingual-gemma2 BAAI | measured MTEB(eng, v2) partial 78.95decision-model estimate p=1.00 +1 | 9.24B | 8K | — | Single GPU | Gemma Terms of Use attribution | 1 | — |
| snowflake-arctic-embed-m Snowflake | measured MTEB(eng, v2) retrieval 57.53MTEB(eng, v2) 61.09 +1 | 109M | 512 | — | Laptop / edge | Apache 2.0 | 1 | — |
| jina-embeddings-v2-base-en jinaai | measured MTEB(eng, v2) retrieval 26.67MTEB(eng, v2) 45.97 +1 | 137M | 8K | — | Laptop / edge | Apache 2.0 | 0 | — |
| Qwen3-VL-Embedding-2B Qwen | author-reported Author reports MMEB-V2 88.5Author reports MMTEB 78.50 +1 | 2.13B | 256K | — | Single GPU | Apache 2.0 | 1 | — |
| Qwen3-VL-Embedding-8B Qwen | author-reported Author reports MMEB-V2 92.2Author reports MMTEB 81.08 +1 | 8.14B | 256K | — | Single GPU | Apache 2.0 | 0 | — |
| gte-large-en-v1.5 Alibaba-NLP | author-reported Author reports MTEB 84.63Author reports LoCo 99.82 +1 | 434M | 8K | — | Laptop / edge | Apache 2.0 | 0 | — |
| Qwen3-Reranker-8B Qwen | author-reported Author reports MLDR 70.19Author reports MTEB-Code 81.22 +1 | 8.19B | 40K | $0.05 | Single GPU | Apache 2.0 | 1 | — |
| Kimi-K3 moonshotai | structural tag: feature-extraction | 2.78T | 1M | $0.23 | Cluster | Kimi K3 (Modified MIT) (k… gated-threshold | 7 | 25.6 s · 59 tok/s |
| Kimi-K2.6 moonshotai | structural tag: feature-extraction | 1.03T | 256K | $0.45 | Cluster | Kimi (Modified MIT) (modi… gated-threshold | 5 | — |
| Kimi-K2.5 moonshotai | structural tag: feature-extraction | 1.03T | 256K | $0.45 | Cluster | Kimi (Modified MIT) (modi… gated-threshold | 4 | — |
| DeepSeek-OCR deepseek-ai | structural tag: feature-extraction | 3.34B | 8K | — | Single GPU | MIT | 1 | — |
| Unlimited-OCR baidu | structural tag: feature-extraction | 3.34B | 32K | — | Single GPU | MIT | 0 | — |
| Qwen3-Reranker-0.6B Qwen | structural tag: sentence-transformers | 596M | 40K | $0.01 | Laptop / edge | Apache 2.0 | 2 | — |
| paraphrase-MiniLM-L6-v2 sentence-transformers | structural decision-model estimate p=1.00tag: sentence-transformers +1 | 22.7M | 512 | $0.0050 | Laptop / edge | Apache 2.0 | 2 | — |
| Kimi-K2.7-Code moonshotai | structural tag: feature-extraction | 1.03T | 256K | $0.67 | Cluster | Kimi (Modified MIT) (modi… gated-threshold | 4 | — |
| bge-reranker-large BAAI | structural name: bgetag: mteb +1 | 560M | 514 | — | Laptop / edge | MIT | 1 | — |
| multi-qa-mpnet-base-dot-v1 sentence-transformers | structural decision-model estimate p=1.00tag: sentence-transformers +1 | 109M | 514 | $0.0050 | Laptop / edge | Not declared review | 2 | — |
| Qwen3-Reranker-4B Qwen | structural tag: sentence-transformers | 4.02B | 40K | $0.03 | Single GPU | Apache 2.0 | 1 | — |
| bge-small-zh-v1.5 BAAI | structural decision-model estimate p=1.00name: bge +1 | 24.0M | 512 | — | Laptop / edge | MIT | 1 | — |
| DeepSeek-OCR-2 deepseek-ai | structural tag: feature-extraction | 3.39B | 8K | — | Single GPU | Apache 2.0 | 0 | — |
| distiluse-base-multilingual-cased-v2 sentence-transformers | structural decision-model estimate p=1.00tag: sentence-transformers +1 | 135M | 512 | — | Laptop / edge | Apache 2.0 | 1 | — |
| bge-large-zh-v1.5 BAAI | structural decision-model estimate p=1.00name: bge +1 | — | 512 | — | — | MIT | 1 | — |
| codebert-base microsoft | structural decision-model estimate p=0.77tag: feature-extraction +1 | — | 514 | — | — | Not declared review | 1 | — |
| distiluse-base-multilingual-cased-v1 sentence-transformers | structural decision-model estimate p=1.00tag: sentence-transformers +1 | 135M | 512 | — | Laptop / edge | Apache 2.0 | 1 | — |
| all-distilroberta-v1 sentence-transformers | structural decision-model estimate p=1.00tag: sentence-transformers +1 | 82.1M | 514 | — | Laptop / edge | Apache 2.0 | 1 | — |
| bart-base facebook | structural tag: feature-extractionpipeline: feature-extraction | 139M | 1K | — | Laptop / edge | Apache 2.0 | 1 | — |
| gte-reranker-modernbert-base Alibaba-NLP | structural name: gtetag: sentence-transformers | 150M | — | — | Laptop / edge | Apache 2.0 | 1 | — |
| LocateAnything-3B nvidia | structural tag: feature-extraction | 3.83B | 32K | — | Single GPU | NVIDIA Open Model License… attribution | 0 | — |
| paraphrase-mpnet-base-v2 sentence-transformers | structural decision-model estimate p=1.00tag: sentence-transformers +1 | 109M | 514 | — | Laptop / edge | Apache 2.0 | 1 | — |
| text2vec-base-chinese shibing624 | structural decision-model estimate p=1.00tag: sentence-transformers +1 | 102M | 512 | $0.0050 | Laptop / edge | Apache 2.0 | 2 | — |
| clip-ViT-B-32-multilingual-v1 sentence-transformers | structural decision-model estimate p=1.00tag: sentence-transformers +1 | 135M | 512 | $0.0050 | Laptop / edge | Apache 2.0 | 2 | — |
| jina-reranker-v2-base-multilingual jinaai | structural tag: sentence-transformers | 278M | 1K | — | Laptop / edge | CC BY-NC 4.0 non-commercial | 0 | — |
| bart-large facebook | structural tag: feature-extractionpipeline: feature-extraction | — | 1K | — | — | Apache 2.0 | 1 | — |
| w2v-bert-2.0 facebook | structural tag: feature-extractionpipeline: feature-extraction | 580M | — | — | Laptop / edge | MIT | 0 | — |
| canine-c google | structural decision-model estimate p=0.75tag: feature-extraction +1 | 132M | — | — | Laptop / edge | Apache 2.0 | 1 | — |
| wavlm-large microsoft | structural tag: feature-extractionpipeline: feature-extraction | — | — | — | — | Not declared review | 0 | — |
| all-roberta-large-v1 sentence-transformers | structural decision-model estimate p=1.00tag: sentence-transformers +1 | 355M | 514 | — | Laptop / edge | Apache 2.0 | 1 | — |
| MiniCPM-o-4_5 openbmb | structural tag: feature-extraction | 9.37B | 40K | — | Single GPU | Apache 2.0 | 0 | — |
| Kimi-VL-A3B-Instruct moonshotai | structural tag: feature-extraction | 16.4B (3B active) | 128K | — | Single high-end GPU | MIT | 0 | — |
| MiniCPM-o-2_6 openbmb | structural tag: feature-extraction | 8.67B | 32K | — | Single GPU | Apache 2.0 | 0 | — |
| NVIDIA-Nemotron-Parse-v1.1 nvidia | structural tag: feature-extraction | 957M | 9K | — | Laptop / edge | NVIDIA Open Model License… attribution | 0 | — |
| MiniCPM-V-4_5 openbmb | structural tag: feature-extraction | 8.70B | 40K | — | Single GPU | Apache 2.0 | 0 | — |
| InternVL3-78B OpenGVLab | structural tag: feature-extraction | 78.4B | — | — | Multi-GPU node | Qwen License (qwen) gated-threshold | 0 | — |
| Kimi-K3-DSpark RadixArk | structural tag: feature-extraction | 2.25B | 1M | — | Single GPU | Not declared review | 0 | — |
| Kimi-VL-A3B-Thinking moonshotai | structural tag: feature-extraction | 16.4B (3B active) | 128K | — | Single high-end GPU | MIT | 0 | — |
| MiniCPM-Llama3-V-2_5 openbmb | structural tag: feature-extraction | 8.54B | 8K | — | Single GPU | Not declared review | 1 | — |
| InternVL3-8B OpenGVLab | structural tag: feature-extraction | 7.94B | — | — | Single GPU | Apache 2.0 (qwen) | 0 | — |
| InternVL3-1B OpenGVLab | structural tag: feature-extraction | 938M | — | — | Laptop / edge | Apache 2.0 (qwen) | 0 | — |
| Kimi-VL-A3B-Thinking-2506 moonshotai | structural tag: feature-extraction | 16.4B (3B active) | 128K | — | Single high-end GPU | MIT | 0 | — |
| system-one-mini DavidHatley | structural tag: feature-extractionpipeline: feature-extraction | 69.3M | — | — | Laptop / edge | Apache 2.0 | 0 | — |
| mojev MoLeMo-Lab | structural tag: feature-extraction | 854M | — | — | Laptop / edge | MIT | 0 | — |
| Lumma-fev-0.6b FrontiersMind | structural tag: feature-extraction | 649M | — | — | Laptop / edge | Apache 2.0 | 0 | — |
| Lumma-fev-0.1b FrontiersMind | structural tag: feature-extraction | 154M | — | — | Laptop / edge | Apache 2.0 | 0 | — |
No model matches every constraint. Loosen one, or switch the evidence filter back to any.
How to read this
- measured a third-party leaderboard row for this model. Settings (agent harness, edit format, prompt) differ between boards; compare within a benchmark, not across.
- author-reported a number the model author published in the card or eval files. Not reproduced by us.
- structural a deterministic signal from the name, config or chat template, or a decision-model estimate. A weak signal: it says the model was built or tagged for the job, not how well it does it. "decision-model estimate" chips are a probability from the Jev decision model, not a measurement.
- Runs on is the memory class of the BF16 weights (method); choosing a class shows that class and smaller. Quantized builds may fit a class lower.
- Price from is the cheapest observed hosted input price, USD per million tokens; models with no observed price are hidden when a ceiling is set. License friction: permissive = none or attribution-only; high = non-commercial.
- Ordering within a tier is the watch-list score (Hub likes, downloads, recency), not a quality verdict.
Subscribe
Follow "Embed text for search"
- /feeds/for/embedding-retrieval.xml(RSS 2.0) · one item per model with measured or author-reported evidence for this job
- /feeds/for/embedding-retrieval.json(JSON Feed 1.1) · same items, JSON Feed
Email alerts are not open yet; the feeds above are the same data. See feeds for the item schema.