Skip to main content
AI21 Labs ↗Released August 22, 2024

Jamba 1.5 Mini

Open Source52B (12B active) parameters

Jamba 1.5 Mini is AI21 Labs's entry in a crowded field. Context window: 0.256K tokens.

Context

256K

Input

$0.20

Key Specifications

🏆

Arena Rank

Not disclosed

📐

Context Window

256K

📥

Input Price

per 1M tokens

$0.20

📤

Output Price

per 1M tokens

$0.40

🧠

Parameters

52B (12B active)

🔓

Open Source

Yes

Best For

Cost-effective long-contextsummarization

About Jamba 1.5 Mini

Jamba 1.5 Mini, developed by AI21 Labs, is a compact hybrid SSM-Transformer model with 52 billion total parameters (12 billion active) and a 256K token context window. The model applies AI21 Labs' hybrid Mamba-Transformer architecture in a smaller, more efficient package designed for cost-effective long-context processing. It handles summarization, document analysis, and RAG tasks with the efficiency advantages of state-space models on long sequences. Priced at $0.20 per million input tokens and $0.40 per million output tokens, it offers affordable access to long-context processing. As an open-source model, Jamba 1.5 Mini can be deployed on more modest hardware compared to its larger sibling. The model targets production applications where processing long documents efficiently at low cost matters more than achieving maximum benchmark scores.

Built byAI21 Labs↗

Pricing per 1M tokens

Input Tokens

$0.20

Output Tokens

$0.40

Compare Jamba 1.5 Mini

See how Jamba 1.5 Mini stacks up against other leading AI models

Jamba 1.5 Mini vs Jamba 1.5 LargeCompare →Jamba 1.5 Mini vs Jamba 1.5 Mini (SSM)Compare →Jamba 1.5 Mini vs Claude Opus 4Compare →Jamba 1.5 Mini vs GPT-4oCompare →Jamba 1.5 Mini vs GPT-o3Compare →

Other AI21 Labs Models

Jamba 1.5 Large
256K
Jamba 1.5 Mini (SSM)
256K

Other Top Models

Claude Opus 4Anthropic
#1
GPT-4oOpenAI
#2
GPT-o3OpenAI
#2
GPT-o1OpenAI
#3
DeepSeek R1DeepSeek
#3
Claude Sonnet 4Anthropic
#3

Explore More

Model Benchmarks →All Models →AI21 Labs Company Profile →

Frequently Asked Questions

What is Jamba 1.5 Mini?▾
Jamba 1.5 Mini, developed by AI21 Labs, is a compact hybrid SSM-Transformer model with 52 billion total parameters (12 billion active) and a 256K token context window. The model applies AI21 Labs' hybrid Mamba-Transformer architecture in a smaller, more efficient package designed for cost-effective long-context processing. It handles summarization, document analysis, and RAG tasks with the efficiency advantages of state-space models on long sequences. Priced at $0.20 per million input tokens and $0.40 per million output tokens, it offers affordable access to long-context processing. As an open-source model, Jamba 1.5 Mini can be deployed on more modest hardware compared to its larger sibling. The model targets production applications where processing long documents efficiently at low cost matters more than achieving maximum benchmark scores.
How much does Jamba 1.5 Mini cost?▾
AI21 Labs charges $0.20 per 1M input tokens for Jamba 1.5 Mini, with output at $0.40. Competitive with other models in its tier.
What is Jamba 1.5 Mini's context window?▾
Jamba 1.5 Mini supports up to 256K tokens per request. A larger context window allows the model to reason over longer inputs, which matters for document analysis, code review, and multi-turn conversations.
Is Jamba 1.5 Mini open source?▾
Yes — AI21 Labs released Jamba 1.5 Mini as open source. That means you're free to deploy it however you want: cloud, on-prem, edge. No API lock-in.
What is Jamba 1.5 Mini best for?▾
AI21 Labs positions Jamba 1.5 Mini for: Cost-effective long-context, summarization. Real-world performance will depend on your specific prompts and data, but these are the intended strengths.