Very long inputs
Choose Rerank v5.0 Fast
Rerank v5.0 Fast reads up to 32k tokens per request, against 4k for Rerank v3.5.
Rerank v3.5 vs Rerank v5.0 Fast: price, context and capabilities
Choose Rerank v5.0 Fast for very long inputs; on the other points the two are level or the data is missing.
In short
- A price comparison is not possible: one of them has no comparable published price.
- Rerank v5.0 Fast reads more per request: 32k tokens against 4k.
Which should you choose?
Each answer below comes straight from the published prices, specifications and independent benchmark results.
Specifications side by side
| Rerank v3.5 | Rerank v5.0 Fast | |
|---|---|---|
| Provider | Cohere | Cohere |
| Category | Reranking | Reranking |
| Status | Legacy | Generally available |
| Headline price | No data | No data |
| Context window | 4k | 32k |
| Max output | No data | No data |
| Released | December 2024 | No data |
| Knowledge cutoff | No data | No data |
| Licence | Proprietary | Proprietary |
| Capabilities | No data | No data |
| API model id | rerank-v3.5 | rerank-v5.0-fast |
Frequently asked questions
Should I choose Rerank v3.5 or Rerank v5.0 Fast?
Choose Rerank v5.0 Fast for very long inputs; on the other points the two are level or the data is missing.
Which is cheaper, Rerank v3.5 or Rerank v5.0 Fast?
A price comparison is not possible: one of them has no comparable published price.
Which has the larger context window, Rerank v3.5 or Rerank v5.0 Fast?
Rerank v5.0 Fast reads more per request: 32k tokens against 4k.
What is the difference between Rerank v3.5 and Rerank v5.0 Fast?
Rerank v3.5 is made by Cohere and Rerank v5.0 Fast by Cohere. A price comparison is not possible: one of them has no comparable published price. Rerank v5.0 Fast reads more per request: 32k tokens against 4k.