AI21 Labs - Mixture of experts - verified
Run AI21 Jamba2 Mini locally
AI21 Jamba2 Mini is a 51.6B mixture-of-experts (12B active) model from AI21 Labs with a 262K-token context window, released under Apache 2.0 (commercial use is allowed); the smallest listed download is 31.2 GB, the Q4_K_M GGUF build from bartowski/ai21labs_AI21-Jamba2-Mini-GGUF.
Official weights: ai21labs/AI21-Jamba2-Mini
- Total params
- 51.6B
- Active params
- 12B
- Context
- 262K
- Smallest download
- 31.2 GB
Fits in
Smallest download at most 85% of memoryDownloads
The official weights and every quantization listed for this model, with file sizes from Hugging Face.
| Download | Format | Size | Repo |
|---|---|---|---|
Official weights | BF16 | 103.1 GB | ai21labs/AI21-Jamba2-Mini |
Q4_K_MSmallest listed | GGUF | 31.2 GB | bartowski/ai21labs_AI21-Jamba2-Mini-GGUF |
Q8_0 | GGUF | 54.8 GB | bartowski/ai21labs_AI21-Jamba2-Mini-GGUF |
GB is decimal: 1 GB = 10^9 bytes. Add memory for the context and the operating system on top of the file size.
Run it
Copied verbatim from the page linked under each command. Server commands from model cards often assume a multi-GPU node; check the flags before you run them.
vLLM
vllm serve "ai21labs/AI21-Jamba2-Mini" --mamba-ssm-cache-dtype float32 --enable-auto-tool-choice --tool-call-parser hermes --enable-prefix-caching --quantization experts_int8Not sure which runtime to use? Compare local LLM runtimes by platform, API and features.
License
- License
- Apache 2.0
apache-2.0 - Commercial use
- Commercial use: YesCommercial use is allowed.
- Gated download
- No. The files download directly.
Worth knowing
SSM-Transformer (Mamba plus attention) MoE with a 256K context window, and the card says best results need vLLM 0.12.0 or newer.
Released (Hugging Face repo creation date).
FAQ
How big is the AI21 Jamba2 Mini download?
The official BF16 weights in ai21labs/AI21-Jamba2-Mini are 103.1 GB. The smallest listed download is 31.2 GB, the Q4_K_M GGUF build from bartowski/ai21labs_AI21-Jamba2-Mini-GGUF. Plan for more memory than the file size: the context (KV cache) and the operating system need room too.
Can I use AI21 Jamba2 Mini commercially?
AI21 Jamba2 Mini is released under Apache 2.0, and commercial use is allowed. The license text is at https://www.apache.org/licenses/LICENSE-2.0.
How much memory do I need for AI21 Jamba2 Mini?
By this directory's rule of thumb it fits in 48 GB and up: the smallest listed download, 31.2 GB, is at most 85% of 48 GB, which leaves room for the context and the operating system.
What is the context length of AI21 Jamba2 Mini?
262,144 tokens. A longer context needs more memory for the KV cache, so you can run with a shorter one when memory is tight.
Sources
Last verified from the pages below.
- HF API (params, license, createdAt) - huggingface.co
- HF file tree - huggingface.co
- Model card - huggingface.co
- GGUF tree (bartowski) - huggingface.co
- GGUF Q8_0 tree (bartowski) - huggingface.co
- Ollama search (no official library entry) - ollama.com
Newsletter
New open weights, every week
One email per week. New model releases, open-weight drops and what they mean for what you build.
Keep exploring
Best local AI models
Our curated picks by hardware tier, with benchmarks and the reasoning.
Model API pricing and specs
First-party prices per 1M tokens when you would rather call an API.
AI benchmark leaderboards
Sourced scores, for when you need to compare quality.
Setup guides
Point your coding tools at the model you picked.