FitMyLLM
AI21/Dense

AI21Jamba 2 3B

Jamba2 3B is an ultra-compact open source model designed to bring enterprise-grade reliability to on-device deployments.

chattool_use
3B
Parameters
256K
Context length
0
Benchmarks
6
Quantizations
0
Architecture
Dense
Released
2026-02-15
Layers
28
KV Heads
1
Head Dim
128
Family
jamba

Quantization Options

Context length:
QuantBitsVRAM @ 16KQuality
Q4_K_M4.89
2.3 GB
2.3 + 0.0 KV
good
Q5_K_S5.57
2.6 GB
2.6 + 0.0 KV
good
Q5_K_M5.7
2.6 GB
2.6 + 0.0 KV
good
Q6_K6.56
3.0 GB
2.9 + 0.0 KV
excellent
Q8_08.5
3.7 GB
3.7 + 0.0 KV
lossless
FP1616
6.5 GB
6.5 + 0.0 KV
lossless

Select your GPU above to see speed estimates and compatibility for each quantization.

Deploying for a team or in production? Size GPUs, cost & scaling in Enterprise →
Ready to run this?Rent by the hour

Rent a GPU and run Jamba 2 3B now

Spin up an A100 / H100 / 4090 in ~60s. Pay by the second. Cancel anytime.

Community Ratings

Loading ratings...

Run this model

Easiest way to get started·Beginners
Docs ↗
curl -fsSL https://ollama.com/install.sh | sh
$# Jamba 2 3B is not in Ollama's library, and we know no GGUF build of it

Search Hugging Face for a GGUF of it (https://huggingface.co/models?sort=trending&search=AI21-Jamba2-3B+GGUF), or use the vLLM tab with the original weights.

Setup guide
>_

Auto-setup with fitmyllm CLI

Detects your GPU, recommends the best model, downloads it, and starts chatting — zero config. Benchmarks your speed and contributes anonymous data to improve predictions.

pip install fitmyllmthen run fitmyllmLearn more
Auto-detect GPULive tok/s in chatSpeed benchmarks9 inference engines

GPUs that can run this model

At Q4_K_M quantization. Sorted by minimum VRAM.

NVIDIA Tesla C2050
3 GB VRAM • 144 GB/s
NVIDIA
NVIDIA Tesla M2050
3 GB VRAM • 148 GB/s
NVIDIA
NVIDIA Tesla S2050
3 GB VRAM • 148 GB/s
NVIDIA

Find the best GPU for Jamba 2 3B

Build Hardware for Jamba 2 3B
Spec sheet

Jamba 2 3B — 3B Dense.

Specifications
Parameters
3B
Architecture
Dense Transformer
Context Length
256K tokens
Capabilities
chat, tool_use
Release Date
2026-02-15
Provider
AI21
Family
jamba
VRAM requirements
QuantBPWVRAMQuality
Q4_K_M4.892.3 GB94%
Q5_K_S5.572.6 GB96%
Q5_K_M5.72.6 GB96%
Q6_K6.562.9 GB97%
Q8_08.53.7 GB100%
FP16166.5 GB100%
Compatible GPUs
30 @ Q4_K_M