Language model
Jamba Mini
ai21
- Released
- January 1, 2026
- Data date
- September 26, 2026
Jamba Mini 1.6 brings AI21's hybrid state-space and Transformer design down to 52 billion total parameters and 12 billion active. It retains a 256K context and grounded generation.
AI21 describes standard serving with at least two 80GB GPUs and ExpertsInt8 on a single 80GB GPU. The card already points to Jamba Mini 1.7 and gives March 5, 2024 as the knowledge cutoff. That boundary matters for any current-fact use case.
Specifications and access
| Specification | Value and source |
|---|---|
| Weights | Open weightsSource |
| Access | APISource |
| Parameters | 52 billionSource |
| Active parameters | 12 billionSource |
| Context window | 256,000 tokensSource |
| Maximum output | 4,096 tokensSource |
| Knowledge Cutoff | August 2024Source |
| License | Jamba Open Model LicenseSource |
| Inputs | TextSource |
| Outputs | TextSource |
| Tool calling | YesSource |
Popularity
9.5 / 100
Sources and data date
| Type | Evidence and data date |
|---|---|
| Additional source | models.dev catalog (retrieved September 26, 2026) |
| Additional source | Hugging Face · AI21 Jamba Mini 1.6 model card · Editorial description reviewed October 2, 2026 |
| Additional source | OpenRouter: Popularity index (derived), CC BY 4.0 |
| Additional source | Vercel AI Gateway: Popularity index (derived), CC BY 4.0 |
| Additional source | Creative Commons Attribution 4.0 |