JA

AI21 Jamba2 3B

JambaGGUF by bartowskiUpdated 2026-01-08
Download on Hugging Face
Parameters3.2B
Context256K
Layers28
Architecturejamba
Smallest quant0.0 GB
Quick answer

AI21 Jamba2 3B needs about 4.8 GB of memory at Q4_K_M (a 1.9 GB file) with an 8K context window, so it fits on a 6 GB GPU. This repository has 25 quants, from 0.0 GB to 6.4 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 9.4 GB.

AI21 Jamba2 3B – compact, enterprise‑ready LLM

AI21 Jamba2 3B is a 3.2B parameter model optimized for on‑device deployment. It offers a 256K context window and is licensed under Apache 2.0, making it suitable for commercial edge applications.

Good at

  • Long‑context instruction following
  • Enterprise RAG workflows
  • Technical document processing

Limits

  • No vision support
  • Requires 256K context window for full capability
  • Large memory for higher quantizations

Best for

  • Edge AI assistants
  • Knowledge‑base search engines
  • Enterprise chatbot deployments

Who it’s for. Developers building AI applications that need reliable instruction following on consumer devices. The model is fully open source and can be deployed on iPhones, Androids, Macs, and PCs.

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.0 GB 2.9 GB
Fits Get file →
IQ2_M1.1 GB 4.0 GB
Fits Get file →
Q2_K1.2 GB 4.1 GB
Fits Get file →
IQ3_XXS1.3 GB 4.1 GB
Fits Get file →
Q2_K_L1.4 GB 4.2 GB
Fits Get file →
IQ3_XS1.4 GB 4.3 GB
Fits Get file →
Q3_K_S1.5 GB 4.3 GB
Fits Get file →
IQ3_M1.5 GB 4.3 GB
Fits Get file →
Q3_K_M1.5 GB 4.4 GB
Fits Get file →
Q3_K_L1.6 GB 4.5 GB
Fits Get file →
Q3_K_XL1.8 GB 4.6 GB
Fits Get file →
IQ4_XS1.8 GB 4.6 GB
Fits Get file →
IQ4_NL1.9 GB 4.7 GB
Fits Get file →
Q4_01.9 GB 4.7 GB
Fits Get file →
Q4_K_S1.9 GB 4.7 GB
Fits Get file →
Q4_K_M1.9 GB 4.8 GB
Fits Get file →
Q4_12.0 GB 4.9 GB
Fits Get file →
Q4_K_L2.1 GB 4.9 GB
Fits Get file →
Q5_K_S2.2 GB 5.1 GB
Fits Get file →
Q5_K_M2.3 GB 5.1 GB
Fits Get file →
Q5_K_L2.4 GB 5.2 GB
Fits Get file →
Q6_K2.6 GB 5.5 GB
Fits Get file →
Q6_K_L2.7 GB 5.6 GB
Fits Get file →
Q8_03.4 GB 6.3 GB
Fits Get file →
BF166.4 GB 9.3 GB
Too big Get file →

For 8 GB VRAM, use Q4_K_M (file size 1.93 GB, estimated 4.8 GB memory). For 12 GB VRAM, Q5_K_L (file size 2.38 GB, estimated 5.2 GB memory) offers higher quality. The BF16 quantization (file size 6.4 GB, estimated 9.3 GB memory) is best for maximum fidelity on high‑end hardware.

What GPU do you need for AI21 Jamba2 3B?

These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is IQ2_M (1.1 GB file, about 4.0 GB in use).
  • 6 GB VRAM: the best quant that fits is Q6_K_L (2.7 GB file, about 5.6 GB in use).
  • 8 GB VRAM: the best quant that fits is Q8_0 (3.4 GB file, about 6.3 GB in use).
  • 12 GB VRAM: the best quant that fits is BF16 (6.4 GB file, about 9.3 GB in use).

Which AI21 Jamba2 3B quant should you choose?

Q4_K_M is the usual starting point for AI21 Jamba2 3B: it balances file size and quality well. Here it is a 1.9 GB file.

The Q8_0 file is 3.4 GB, about 1.8 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.

If memory is tight, the smallest file is the first one in the table at 0.0 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturejamba
Parameters3.2B
Layers28
Trained context262,144 tokens
KV cache (fp16)280 KB per token

Source

Quantized bybartowski
FamilyJamba
Licenseapache-2.0
Downloads (HF)864
Last updated2026-01-08

Frequently asked questions about AI21 Jamba2 3B

How much VRAM does AI21 Jamba2 3B need?

About 4.8 GB at Q4_K_M with an 8K context window. That is the 1.9 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run AI21 Jamba2 3B on an 8 GB GPU?

Yes, with Q8_0: the 3.4 GB file needs about 6.3 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of AI21 Jamba2 3B GGUF?

It depends on the quant: file is 0.0 GB, IQ2_M is 1.1 GB, Q2_K is 1.2 GB, IQ3_XXS is 1.3 GB, Q2_K_L is 1.4 GB, IQ3_XS is 1.4 GB and 19 more.

What context length does AI21 Jamba2 3B support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 2.3 GB at 8K and 9.4 GB at 32K.

Where can I download AI21 Jamba2 3B GGUF?

The files are on Hugging Face in the bartowski/ai21labs_AI21-Jamba2-3B-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does AI21 Jamba2 3B use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Can I run AI21 Jamba2 3B on 8 GB VRAM?

Yes, the Q4_K_M quantization uses about 4.8 GB memory at 8 k context, fitting within an 8 GB VRAM budget.

Does the model support image inputs?

No, the model only supports text; it has no vision capabilities.

What license applies to this model?

The model is released under the Apache 2.0 license, allowing commercial use.

Other GGUFs of AI21-Jamba2-3B

Similar models

Browse all →

More from bartowski

All by bartowski →

More Jamba models

All Jamba →

Other tiny models

All tiny models →