As of: 2026-11-13 13:49 CET.
Researched and written by AI agents (Claude Opus 5.5 in Claude Code), with a human editor. How we work: https://postcutoff.com/about/
Canonical page: https://postcutoff.com/m/cohere-rerank-4/

# Cohere Rerank 4 (Pro / Fast)

Cohere, Rerank family. Status: current. Type: embedding model. Released: 2025-12-11.

## How to call it

| Provider | Model id | Endpoint | Docs |
|---|---|---|---|
| Cohere API | `rerank-v4.0-pro` | https://api.cohere.com/v2/rerank | https://docs.cohere.com/docs/models |
| Cohere API (fast) | `rerank-v4.0-fast` | https://api.cohere.com/v2/rerank |  |
| OpenRouter | `cohere/rerank-4-pro` |  | https://openrouter.ai/cohere/rerank-4-pro |

## Pricing

- Price not published.
- Last checked: 2026-09-29

## Specs

- Input: text
- Output: text
- Context window: 32,000 tokens
- Max output: not published
- Open weights: no
- Licence: proprietary

## Where it stands

- Newer embedding models in the registry since its release: 4:
  - EmbeddingGemma 2 (2026-10-06) https://postcutoff.com/m/embeddinggemma-2/
  - Cohere Embed 5 Fast (2026-09-30) https://postcutoff.com/m/cohere-embed-v5-fast/
  - Cohere Embed 5 Pro (2026-09-30) https://postcutoff.com/m/cohere-embed-v5-pro/
  - Gemini Embedding 2 (2026-04-22) https://postcutoff.com/m/gemini-embedding-2/
- Before it from Cohere: Cohere Embed v4 (2025-04) https://postcutoff.com/m/cohere-embed-v4/
- After it from Cohere: Cohere Embed 5 Pro (2026-09-30) https://postcutoff.com/m/cohere-embed-v5-pro/
- Closed weights; access is hosted only.

## What stands out

- 32K-context reranking: Rerank window grew from 4K (v3.5) to 32K tokens, so whole long documents can be scored. Source: https://docs.cohere.com/docs/models
- Pro / Fast tiers: Two variants: pro for best accuracy, fast for latency-sensitive search. Source: https://docs.cohere.com/docs/models

## Notes

Reranker (scores query-document relevance). Previous: rerank-v3.5 (Bedrock cohere.rerank-v3-5:0). Release date from third-party listing; pricing not verified (OpenRouter ~$0.0025/search reported, not checked).

Second-stage reranking for RAG and enterprise search.

```bash
curl https://api.cohere.com/v2/rerank -H "Authorization: Bearer $CO_API_KEY" -H "Content-Type: application/json" \
 -d '{"model":"rerank-v4.0-pro","query":"capital of France","documents":["Paris is the capital of France.","Berlin is in Germany."]}'
```

Sources: https://docs.cohere.com/docs/models

Model page: https://postcutoff.com/m/cohere-rerank-4/