Wednesday, October 28, 2026 🏢 AI Companies Hub RSS About Contact Admin
POPULAR BEATS: Generative AI LLMs & NLP Autonomous Agents Robotics & Hardware Enterprise AI AI Ethics & Policy 🏢 All AI Companies

Jamba 1.5 Large

SSM-Transformer (Mamba + Attention) MoE
Official Source & Paper
AI21 Labs logo AI21 Labs 📄 Open Weights (Open Jamba License) 🌍 Israel (Global) 🗓️ Effective Date: 22.08.2024

World-first production-scale hybrid architecture intertwining structured state-space layers (Mamba) with Transformer attention for 256k efficiency.

Model Size / Parameters 398B (94B active) Architecture: SSM-Transformer (Mamba + Attention) MoE
Context Window (Input) 256k tokens Token Capacity
Max Output Tokens 16k tokens Single Generation Limit
Supported Modalities Text, Code Native Multimodal Ingestion
License & Access Open Weights (Open Jamba License) Open Weights vs. API
Release Date August 2024 Effective Date: 22.08.2024

Verified Benchmark Results

Standardized academic and industry evaluation benchmarks (Effective Date: 22.08.2024).

MMLU-Pro / 5-Shot General Intellect
81.2%
SWE-bench Verified Software Engineering
36.5%
MATH 500 / AIME Mathematical Reasoning
74.8%
GPQA Diamond Doctoral Science
45.0%
HumanEval / Python Algorithm Synthesis
79.5%

Pricing & Cost Efficiency

Official API token pricing and self-hosted licensing terms.

Input Pricing (Prompt)
$2.00 / 1M tokens
Per 1 million processed tokens
Output Pricing (Generation)
$8.00 / 1M tokens
Per 1 million generated tokens
Verified Datasheet: Source: AI21 Labs Research Documentation • Effective Date: 22.08.2024
← Back to All AI Model Datasheets