TypeSafe AIDecision modelAvailable via API
Jev 1.14
Summary
Jev 1.14 is a pinned API version of TypeSafe's decision model. It scores supplied options, assertions and scales against shared context. With image support enabled, a Box question can return the location of a signature on a scan.
TypeSafe lists default limits of 64k tokens per request and 32k per question. Images and longer context use separate experimental headers. The rate of $0.042 per million tokens applies to input.
Similar models
More models of the same kind as Jev 1.14, chosen by model line, developer, size, price and release date.
- JevTypeSafe AI · same developer
- AutoTrust JEV-27B-VLAutoTrust AI · released around the same time
- ClefCloudflare · released around the same time
- Clef-FlashCloudflare · released around the same time
- pplx-decider-v1.1-27bPerplexity · released around the same time
- AutoTrust JEV-27BAutoTrust AI · released around the same time
- Drex MaxNace.AI · released around the same time
- pplx-decider-flash-v1.2Perplexity · released around the same time
All specifications
Input and output
- Input
- Text. PNG and JPEG images and Box questions require the TypeSafe-Danger-Zone-Images: true header.
- Source: Input
Watermarking and provenance
AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.
- AI watermarking
- Unknown
- no source
- Watermarking technique
- Not known
- no source
Licence and access
- Shutdown
- None announced
- no source
API
Further details
- Question types
- Choice for answer options, Score for scales, Noul for yes/no, and Box for image regions.
- Source: Question types
- Input limits
- TypeSafe documents 64k tokens per request and 32k per question by default. With the experimental long-context header, it lists 262k tokens for both limits.
- Source: Input limits