This model is deprecated. We still benchmark its performance, but less often: every 18 hours instead of every 6.
AI21 Labs has launched a newer release, Jamba 1.7 Large. We suggest considering it instead.
Jamba 1.6 Large: Intelligence, Performance & Price Analysis
Model summary
Performance
Verbosity
Jamba 1.6 Large supports text input, outputs text, and has a 256k tokens context window.
| Reasoning | No |
|---|---|
| Input modality | Supports: text |
| Output modality | Supports: text |
| Context window | 256k ~384 A4 pages of size 12 Arial font |
| Total parameters | 398B |
| Active parameters | 94B Number of parameters active per token during inference |
| License | Jamba Open Model License Agreement |
| Model weights | Hugging Face |
Metrics are compared against models of the same class:
- Non-reasoning models → compared only with other non-reasoning models
- Reasoning models → compared across both reasoning and non-reasoning
- Open weights models → compared only with other open weights models of the same size class:
- Tiny: ≤4B parameters
- Small: 4B–40B parameters
- Medium: 40B–150B parameters
- Large: >150B parameters
- Proprietary models → compared across proprietary and open weights models of the same price range, using a blended 3:1 input/output price ratio:
- <$0.15 per 1M tokens
- $0.15–$1 per 1M tokens
- >$1 per 1M tokens
Time per Task
Intelligence
Artificial Analysis Intelligence Index
Artificial Analysis Intelligence Index by Open Weights / Proprietary
Intelligence Index vs. Cost per Task
Openness Index
Artificial Analysis Openness Index: Score
Cost
Cost per Intelligence Index Task
Cost to Run Artificial Analysis Intelligence Index
Pricing: Cache Hit, Input, and Output
Context Window
Context Window
Model Size (Open Weights Models Only)
Model Size: Total and Active Parameters
Further details
Weights | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| Jamba 1.6 Large | - | 398B 94B active at inference time | 256k | - | - | - | - |
More from AI21 Labs
Frequently Asked Questions
Common questions about Jamba 1.6 Large
Jamba 1.6 Large was released on March 6, 2025.
Jamba 1.6 Large was created by AI21 Labs.
No, Jamba 1.6 Large is not a reasoning model. It provides direct responses without extended chain-of-thought reasoning.
Jamba 1.6 Large supports text input.
Jamba 1.6 Large supports text output.
No, Jamba 1.6 Large does not support image input. It can only process text.
No, Jamba 1.6 Large is not multimodal. It only supports text input.
Jamba 1.6 Large has a context window of 256k tokens. This determines how much text and conversation history the model can process in a single request.
Yes, Jamba 1.6 Large is open weights. The model weights are publicly available and can be downloaded for self-hosting.
Jamba 1.6 Large has 398B parameters (94B active).
Jamba 1.6 Large is a Mixture of Experts (MoE) model with 398B total parameters, but only 94B active parameters are used during inference.
Jamba 1.6 Large is released under the Jamba Open Model License Agreement license. Commercial use is allowed with restrictions. View license
Jamba 1.6 Large is an open weights model that can be self-hosted. View providers
Jamba 1.6 Large is an open weights model that can be downloaded and self-hosted. Compare providers