AI21 Jamba Reasoning 3B — Hardware Requirements & GPU Compatibility
ChatReasoning
2.9K downloads
133 likesSpecifications
- Publisher
- AI21 Labs
- Parameters
- 3.2B
- Architecture
- JambaForCausalLM
- Context Length
- 262,144 tokens
- Vocabulary Size
- 65,536
- Release Date
- 2025-10-08
- License
- Apache 2.0
Get Started
HuggingFace
How Much VRAM Does AI21 Jamba Reasoning 3B Need?
Select a quantization to see compatible GPUs below.
| Quantization | Bits | VRAM | + Context | File Size | Quality |
|---|---|---|---|---|---|
| BF16 | 16.00 | 6.7 GB | 10.4 GB | 6.39 GB | Brain floating point 16 — preferred for training |
Which GPUs Can Run AI21 Jamba Reasoning 3B?
BF16 · 6.7 GBAI21 Jamba Reasoning 3B (BF16) requires 6.7 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 9+ GB is recommended. Using the full 262K context window can add up to 3.7 GB, bringing total usage to 10.4 GB. 35 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti, NVIDIA GeForce RTX 3080.
Runs great
— Plenty of headroom
Which Devices Can Run AI21 Jamba Reasoning 3B?
BF16 · 6.7 GB33 devices with unified memory can run AI21 Jamba Reasoning 3B, including NVIDIA DGX H100, NVIDIA DGX A100 640GB, MacBook Air 13" M3 (8 GB).
Runs great
— Plenty of headroomDecent
— Enough memory, may be tightRelated Models
Frequently Asked Questions
- How much VRAM does AI21 Jamba Reasoning 3B need?
AI21 Jamba Reasoning 3B requires 6.7 GB of VRAM at BF16. Full 262K context adds up to 3.7 GB (10.4 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 3.2B × 16 bits ÷ 8 = 6.4 GB
KV Cache + Overhead ≈ 0.3 GB (at 2K context + ~0.3 GB framework)
KV Cache + Overhead ≈ 4 GB (at full 262K context)
VRAM usage by quantization
BF166.7 GBBF16 + full context10.4 GB- Can I run AI21 Jamba Reasoning 3B on a Mac?
AI21 Jamba Reasoning 3B requires at least 6.7 GB at BF16, which exceeds the unified memory of most consumer Macs. You would need a Mac Studio or Mac Pro with a high-memory configuration.
- Can I run AI21 Jamba Reasoning 3B locally?
Yes — AI21 Jamba Reasoning 3B can run locally on consumer hardware. At BF16 quantization it needs 6.7 GB of VRAM. Popular tools include Ollama, LM Studio, and llama.cpp.
- How fast is AI21 Jamba Reasoning 3B?
At BF16, AI21 Jamba Reasoning 3B can reach ~434 tok/s on AMD Instinct MI300X. On NVIDIA GeForce RTX 4090: ~98 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: AMD Instinct MI300X → 5300 ÷ 6.7 × 0.55 = ~434 tok/s
Estimated speed at BF16 (6.7 GB)
AMD Instinct MI300X~434 tok/sNVIDIA GeForce RTX 4090~98 tok/sNVIDIA H100 SXM~324 tok/sAMD Instinct MI250X~268 tok/sReal-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
- What's the download size of AI21 Jamba Reasoning 3B?
At BF16, the download is about 6.39 GB.
- Which GPUs can run AI21 Jamba Reasoning 3B?
35 consumer GPUs can run AI21 Jamba Reasoning 3B at BF16 (6.7 GB). Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT, AMD Radeon RX 7600. 27 GPUs have plenty of headroom for comfortable inference.
- Which devices can run AI21 Jamba Reasoning 3B?
33 devices with unified memory can run AI21 Jamba Reasoning 3B at BF16 (6.7 GB), including Mac Mini M4 (16 GB), Mac Mini M4 (32 GB), Mac Mini M4 Pro (24 GB), Mac Mini M4 Pro (48 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.