JA

AI21 Jamba Large 1.7

JambaGGUF by bartowskiUpdated 2025-10-08
Download on Hugging Face
Parameters399B
Context256K
Layers72
Architecturejamba
Smallest quant0.1 GB
Quick answer

AI21 Jamba Large 1.7 needs about 260.3 GB of memory at Q4_K_M (a 240.4 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 27 quants, from 0.1 GB to 423.6 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 77.3 GB.

AI21 Jamba Large 1.7 is a 399B-parameter Jamba model in GGUF format, quantized by bartowski. Its trained context length is 256K tokens. The repository has 27 quant files, from 0.1 GB to 423.6 GB (Q8_0).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.1 GB 19.9 GB
Too big Get file →
IQ1_S81.9 GB (3 parts) 101.7 GB
Too big Get file →
IQ1_M85.9 GB (3 parts) 105.7 GB
Too big Get file →
IQ2_XXS96.4 GB (3 parts) 116.2 GB
Too big Get file →
IQ2_XS110.7 GB (3 parts) 130.5 GB
Too big Get file →
IQ2_S111.1 GB (3 parts) 131.0 GB
Too big Get file →
IQ2_M125.8 GB (4 parts) 145.6 GB
Too big Get file →
Q2_K138.1 GB (4 parts) 157.9 GB
Too big Get file →
Q2_K_L138.6 GB (4 parts) 158.4 GB
Too big Get file →
IQ3_XXS155.8 GB (4 parts) 175.6 GB
Too big Get file →
IQ3_XS163.1 GB (5 parts) 182.9 GB
Too big Get file →
Q3_K_S171.8 GB (5 parts) 191.6 GB
Too big Get file →
IQ3_M179.6 GB (5 parts) 199.4 GB
Too big Get file →
Q3_K_M180.5 GB (5 parts) 200.3 GB
Too big Get file →
Q3_K_L188.5 GB (5 parts) 208.3 GB
Too big Get file →
Q3_K_XL188.9 GB (5 parts) 208.8 GB
Too big Get file →
IQ4_XS212.1 GB (6 parts) 232.0 GB
Too big Get file →
IQ4_NL224.5 GB (6 parts) 244.4 GB
Too big Get file →
Q4_0228.2 GB (6 parts) 248.0 GB
Too big Get file →
Q4_K_S231.9 GB (6 parts) 251.7 GB
Too big Get file →
Q4_K_M240.4 GB (7 parts) 260.3 GB
Too big Get file →
Q4_K_L240.8 GB (7 parts) 260.7 GB
Too big Get file →
Q4_1249.3 GB (7 parts) 269.2 GB
Too big Get file →
Q5_K_S274.2 GB (8 parts) 294.0 GB
Too big Get file →
Q5_K_M282.4 GB (8 parts) 302.2 GB
Too big Get file →
Q6_K327.1 GB (9 parts) 346.9 GB
Too big Get file →
Q8_0423.6 GB (11 parts) 443.4 GB
Too big Get file →

What GPU do you need for AI21 Jamba Large 1.7?

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: no quant fits at an 8K context.
  • 6 GB VRAM: no quant fits at an 8K context.
  • 8 GB VRAM: no quant fits at an 8K context.
  • 12 GB VRAM: no quant fits at an 8K context.
  • 16 GB VRAM: no quant fits at an 8K context.
  • 24 GB VRAM: the best quant that fits is the main file (0.1 GB file, about 19.9 GB in use).
  • 32 GB VRAM: the best quant that fits is the main file (0.1 GB file, about 19.9 GB in use).
  • 48 GB VRAM: the best quant that fits is the main file (0.1 GB file, about 19.9 GB in use).

Which AI21 Jamba Large 1.7 quant should you choose?

Q4_K_M is the usual starting point for AI21 Jamba Large 1.7: it balances file size and quality well. Here it is a 240.4 GB file.

The Q8_0 file is 423.6 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 0.1 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
Parameters399B (approximate, mixture of experts)
Layers72
Trained context262,144 tokens
KV cache (fp16)2304 KB per token

Source

Quantized bybartowski
FamilyJamba
Licenseother
Downloads (HF)342
Last updated2025-10-08

Frequently asked questions about AI21 Jamba Large 1.7

How much VRAM does AI21 Jamba Large 1.7 need?

About 260.3 GB at Q4_K_M with an 8K context window. That is the 240.4 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 Jamba Large 1.7 on an 8 GB GPU?

Not at an 8K context. The smallest file needs about 19.9 GB, so you would need a larger GPU or CPU offloading.

What is the file size of AI21 Jamba Large 1.7 GGUF?

It depends on the quant: file is 0.1 GB, IQ1_S is 81.9 GB, IQ1_M is 85.9 GB, IQ2_XXS is 96.4 GB, IQ2_XS is 110.7 GB, IQ2_S is 111.1 GB and 21 more.

What context length does AI21 Jamba Large 1.7 support?

The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 19.3 GB at 8K and 77.3 GB at 32K.

Where can I download AI21 Jamba Large 1.7 GGUF?

The files are on Hugging Face in the bartowski/ai21labs_AI21-Jamba-Large-1.7-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 Jamba Large 1.7 use?

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

Other GGUFs of AI21-Jamba-Large-1.7

Similar models

Browse all →

More from bartowski

All by bartowski →

More Jamba models

All Jamba →

Other large models

All large models →