Jamba Mini

Jamba Mini is AI21 Labs' efficient, lightweight hybrid SSM-Transformer model for a wide range of tasks. It is part of the Jamba line, following Jamba Large, and is positioned as a smaller, more cost-effective option. Jamba Mini has a context window of 256,000 tokens and a maximum output of 4,096 tokens. It accepts and returns text only. It supports tool calling but not reasoning mode. The weights are open. Pricing starts at $0.20 per million input tokens and $0.40 per million output tokens. The knowledge cutoff is August 22, 2024.

Against Jamba Large: input $2.00 to $0.20, output $8.00 to $0.40.

• ·256k token context window
• ·4,096 token max output
• ·Tool calling supported
• ·Open weights
• ·Priced from $0.20/M input

• Reach for it for processing long documents
• Reach for it for tool-calling workflows
• Reach for it for text-only generation tasks
• Reach for it for cost-sensitive deployments

1 host serve this model at the price above · the maker's documentation

The registry keeps the six most recent versions of a line. Older ones stay in the file and can be surfaced without a rebuild.

What is the context window of Jamba Mini?

Jamba Mini has a context window of 256,000 tokens, allowing it to handle very long inputs in a single pass.

Does Jamba Mini support tool calling?

Yes, Jamba Mini supports tool calling, which lets it interact with external functions or APIs. It does not support reasoning mode.

What is the price for Jamba Mini?

Pricing starts at $0.20 per million input tokens and $0.40 per million output tokens. The weights are open, so you can also self-host.