Cohere Embed 5 Fast
Latency/cost tier for live queries; ~2.4x Pro throughput; ViDoRe V3 84.5 (Cohere). Output is vectors (modality_out text used as placeholder). Bedrock/Foundry model ids not verified.
- Context window
- 128,000 tokens
- Input
- text, image
- Output
- text
- License
- proprietary
- Pricing
- input: $0.08 (per 1M text tokens (images $0.40 per 1M tokens)) source
- Verified
- 2026-10-01
How to call it
| Provider | Model id | Endpoint / URL | Docs |
|---|---|---|---|
| Cohere API | embed-v5.0-fast | https://api.cohere.com/v2/embed | docs |
| Microsoft Foundry | — | — | — |
| Amazon SageMaker | — | — | — |
| Cohere Model Vault | — | — | — |
Notable capabilities (2)
- Shared embedding space across tiers: Embed 5 Pro and Fast vectors are interoperable: index with Pro, query with Fast without re-indexing. source
- Multimodal 128K embeddings: Text, image and fused text+image input, 100+ languages, Matryoshka dims 256-2048. source
curl https://api.cohere.com/v2/embed -H "Authorization: Bearer $CO_API_KEY" -H "Content-Type: application/json" \
-d '{"model":"embed-v5.0-fast","input_type":"search_query","embedding_types":["float"],"texts":["hello world"]}'
Timeline entry
- Cohere releases Embed 5 Pro and Fast: multimodal embeddings that share one vector space, top ViDoRe V3 ★★
On Sept 30, 2026 Cohere released Embed 5 in two tiers, Pro (embed-v5.0-pro) and Fast (embed-v5.0-fast). Both take text, images and mixed inputs in 100+ languages with a 128K-token context and share one embedding space, so teams can index with Pro and query with Fast without re-indexing. Cohere…
Other Cohere models
Cohere Embed 5 Pro · Command A+ · Cohere Rerank 4 (Pro / Fast) · Cohere Embed v4 · Command A (03-2025) and variants