Embed 5 Fast
Cohere's lightweight embedding model optimized for low latency and high throughput in interactive search, agent loops, and high-volume retrieval. Embed 5 Fast supports text, images, and mixed content…
vercelAIGatewayCatalog/cohere/embed-v5.0-fast
Cohere's lightweight embedding model optimized for low latency and high throughput in interactive search, agent loops, and high-volume retrieval. Embed 5 Fast supports text, images, and mixed content, over 100 languages, and a 128,000-token context window, with output dimensions from 256 to 2,048. It shares an embedding space with Embed 5 Pro.
Specifications & capabilities
- Input
- text
- Output
- text
- Context window · tokens
- 128000
Evaluations
1–4 / 4
68.175%
- source id
- mteb
- leaderboard name
- MTEB Summarization
- dimension
- embedding
- metric
- Average task score
- rank
- 36
- sample size
- 1
- aggregation scope
- task_count
- 1
- sample_size
- 1
- subset_count
- 1
- aggregation_method
- mean-of-task-means
- model_revision_count
- 1
- source model id
- Cohere/Cohere-embed-v5.0-fast
Original result ↗64.698%
- source id
- mteb
- leaderboard name
- MTEB Retrieval
- dimension
- embedding
- metric
- Average task score
- rank
- 261
- sample size
- 29
- aggregation scope
- subsets
- ar
- bn
- de
- default
- en
- es
- fa
- fi
- fr
- hi
- id
- ja
- ko
- ru
- sw
- te
- th
- yo
- zh
- task_count
- 12
- task_names
- AppsRetrieval
- ChatDoctorRetrieval
- DS1000Retrieval
- FinQARetrieval
- FinanceBenchRetrieval
- FreshStackRetrieval
- HC3FinanceRetrieval
- HumanEvalRetrieval
- MBPPRetrieval
- MIRACLRetrievalHardNegatives
- SWEbenchCodeRetrieval
- WikiSQLRetrieval
- sample_size
- 29
- subset_count
- 19
- subset_sample
- ar
- bn
- de
- default
- en
- es
- fa
- fi
- fr
- hi
- id
- ja
- ko
- ru
- sw
- te
- th
- yo
- zh
- aggregation_method
- mean-of-task-means
- model_revision_count
- 1
- source model id
- Cohere/Cohere-embed-v5.0-fast
Original result ↗62.149%
- source id
- mteb
- leaderboard name
- MTEB Overall
- dimension
- embedding
- metric
- Average task score
- rank
- 263
- sample size
- 66
- aggregation scope
- splits
- all
- dentistry_and_oral_health
- dermatology
- dev
- gastroenterology
- genetics
- neuroscience_and_neurology
- orthopedic_surgery
- otorhinolaryngology
- plastic_surgery
- psychiatry_and_psychology
- pulmonology
- test
- subsets
- ar
- bn
- de
- default
- en
- es
- fa
- fi
- fr
- hi
- id
- ja
- ko
- ru
- sw
- te
- th
- yo
- zh
- task_count
- 17
- task_names
- AILACasedocs
- AILAStatutes
- AppsRetrieval
- CUREv1
- ChatDoctorRetrieval
- DS1000Retrieval
- FinQARetrieval
- FinanceBenchRetrieval
- FreshStackRetrieval
- HC3FinanceRetrieval
- HumanEvalRetrieval
- LegalQuAD
- LegalSummarization
- MBPPRetrieval
- MIRACLRetrievalHardNegatives
- SWEbenchCodeRetrieval
- WikiSQLRetrieval
- sample_size
- 66
- subset_count
- 19
- subset_sample
- ar
- bn
- de
- default
- en
- es
- fa
- fi
- fr
- hi
- id
- ja
- ko
- ru
- sw
- te
- th
- yo
- zh
- aggregation_method
- mean-of-task-means
- model_revision_count
- 1
- source model id
- Cohere/Cohere-embed-v5.0-fast
Original result ↗52.998%
- source id
- mteb
- leaderboard name
- MTEB Other
- dimension
- embedding
- metric
- Average task score
- rank
- 413
- sample size
- 36
- aggregation scope
- splits
- all
- dentistry_and_oral_health
- dermatology
- gastroenterology
- genetics
- neuroscience_and_neurology
- orthopedic_surgery
- otorhinolaryngology
- plastic_surgery
- psychiatry_and_psychology
- pulmonology
- test
- task_count
- 4
- task_names
- AILACasedocs
- AILAStatutes
- CUREv1
- LegalQuAD
- sample_size
- 36
- subset_count
- 4
- aggregation_method
- mean-of-task-means
- model_revision_count
- 1
- source model id
- Cohere/Cohere-embed-v5.0-fast
Original result ↗Explore more models