Jamba 1.5 Mini

Overview

AI21 Labs' hybrid SSM-Transformer with MoE routing, activating 12B of 52B parameters. Delivers a verified 256k context window but ships under AI21's non-OSI Jamba license.

When to pick this model

• Long-document workflows that genuinely use 200k+ tokens
• Multilingual chat across the 9 supported languages
• Benchmarking SSM-Transformer hybrids against pure attention models
• Use cases where the Jamba license terms are acceptable

VRAM requirements by quantization

VRAM figures include model weights plus a typical 8k KV cache and ~600 MB runtime overhead (Ollama / llama.cpp baseline). Add headroom for higher context lengths.

Strengths

• Effective 256k context (86% on RULER)
• Unique SSM-Transformer hybrid architecture
• Strong throughput vs. dense models of similar capability
• Solid 9-language coverage

Limitations

• Custom Jamba license is not OSI-approved
• Partial llama.cpp support complicates local deployment
• Superseded by Jamba 1.6 and 1.7
• Smaller fine-tune ecosystem than Llama or Qwen

Architecture & training

Architecture: Hybrid SSM-Transformer (Mamba+Attention) + MoE · 52B/12B active

Training: 256k effective ctx (86% at 256k RULER).

Quick start

# HuggingFace : ai21labs/AI21-Jamba-Mini-1.5

Or use the open-source MCP server to query this model from Claude Desktop, Cursor, or any MCP-compatible client.