Zamba2 1.2B Instruct — Hardware Requirements & GPU Compatibility
Chat
Zamba2 1.2B Instruct is a 1.2B-parameter open language model from Zyphra. It supports a context window of up to 4,096 tokens. At Q4_K_M it needs about 1.67 GB of VRAM — see which GPUs and Macs can run it below.
Runs great on a 24 GB GPU
Runs from 8 GB (Q4_K_M · 4K)Best on 24 GB (Q4_K_M · ~392 tok/s · 4K)Macs from 8 GB (Q4_K_M · ~43 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Zamba2 1.2B Instruct
One recommended way. Every other method, OS and quant is in the step-by-step guide.Each way opens a step-by-step guide for your OS and quant.
Recommended
ollama run hf.co/1bit-MONSTER/Zamba2-1.2B-instruct-GGUFQ4_K_M fits fully on the RTX 4090 · 21.7 GB free · 4K context
Download: Zamba2-1.2B-instruct-Q4_K_M.gguf0.73 GB1bit-MONSTER
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3080 Ti (12 GB) (9 GB headroom) · Q4_K_M
- Generation speed
- ~355 tok/s
- generation speed
- Cost per 1M output tokens
- $0.096
- per 1M output tokens
How Much VRAM Does Zamba2 1.2B Instruct Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 1.4 GB | — | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 4K context · ~452 tok/s | |
| 1.5 GB | — | 3-bit medium — 3-bit medium quantization | Fits: 4K context · ~428 tok/s | |
| 1.7 GB | — | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 4K context · ~392 tok/s | |
| 1.8 GB | — | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 4K context · ~364 tok/s | |
| 1.9 GB | — | 6-bit, very good — 6-bit quantization, very good quality | Fits: 4K context · ~338 tok/s | |
| 2.1 GB | — | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 4K context · ~305 tok/s | |
| 3.4 GB | — | 16-bit · bfloat — Brain floating point 16 — preferred for training | Fits: 4K context · ~194 tok/s |
Selected Q4_K_M
4-bit medium quantization2.3GBat 4K context
- Weights
- 0.7 GB
- KV cache
- 1.3 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 21.7 GB free · ~392 tok/s
≥50 tok/s needs ≥128 GB/s · RTX 5090 ~698, RTX 3090 Ti ~392
Which Hardware Can Run Zamba2 1.2B Instruct?
Q4_K_M · 1.7 GBZamba2 1.2B Instruct (Q4_K_M) requires 1.7 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 3+ GB is recommended. Using the full 4K context window can add up to 0.6 GB, bringing total usage to 2.3 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run Zamba2 1.2B Instruct, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB).
≤ 11 GB
≤ 11 GB tier →fits · 5.7 GB free · up to 4K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~296Simulate output on RTX 3080RTX 2080 Ti~240Simulate output on RTX 2080 TiRTX 3070 Ti~237Simulate output on RTX 3070 TiM3*8 GB~43Simulate output on iMac 24" M3 (8 GB)Orin Nano~40Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~29Simulate output on iPhone 1720 GB
fits · 17.7 GB free · up to 4K ctx
1 GPU
RX 7900 XT~287Simulate output on RX 7900 XT32–40 GB
fits · 29.7 GB free · up to 4K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~698Simulate output on RTX 5090M5 Max36 GB~193Simulate output on Mac Studio M5 Max (36 GB)M532 GB~64Simulate output on MacBook Air 13" M5 (32 GB)Orin~80Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~53Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| MiniCPM V 4.6 ThinkingOpenBMB | — | 1.5 GB | ~580 | same size |
| Llama 3.2 1B InstructMeta | — | 2.3 GB | ~799 | same size |
| LFM2.5 1.2B ThinkingLiquid AI | — | 1.2 GB | ~612 | same size |
| LFM2 1.2BLiquid AI | — | 1.2 GB | ~612 | same size |
| T5gemma L L Ul2 ITGoogle | — | 4.3 GB | ~240 | same size |
| OLMo 1B HFAllen AI | — | 1.3 GB | ~516 | same size |
Specifications
- Publisher
- Zyphra
- Parameters
- 1.2B
- Architecture
- Zamba2ForCausalLM
- Context Length
- 4,096 tokens
- Vocabulary Size
- 32,000
- Release Date
- 2024-09-19
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does Zamba2 1.2B Instruct need?
Zamba2 1.2B Instruct requires 1.7 GB of VRAM at Q4_K_M, or 3.4 GB at BF16. Full 4K context adds up to 0.6 GB (2.3 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 1.2B × 4.8 bits ÷ 8 = 0.7 GB
KV Cache + Overhead ≈ 1 GB (at 2K context + ~0.3 GB framework)
Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.
KV Cache + Overhead ≈ 1.6 GB (at full 4K context)
VRAM usage by quantization
What's the best quantization for Zamba2 1.2B Instruct?
For Zamba2 1.2B Instruct, Q4_K_M (1.7 GB) offers the best balance of quality and VRAM usage. Q5_K_M (1.8 GB) provides better quality if you have the VRAM. The smallest option is Q2_K at 1.4 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Zamba2 1.2B Instruct on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run Zamba2 1.2B Instruct. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 1.7 GB of usable unified memory (RAM minus macOS overhead).
Can I run Zamba2 1.2B Instruct locally?
Yes — Zamba2 1.2B Instruct can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Zamba2 1.2B Instruct?
At Q4_K_M, Zamba2 1.2B Instruct can reach ~2874 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~392 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = (bandwidth GB/s ÷ model GB) × efficiency
Example: NVIDIA B200 → 8000 ÷ 1.67 × 0.65 = ~3114 tok/s
Estimated speed at Q4_K_M (1.7 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Zamba2 1.2B Instruct?
At Q4_K_M, the download is about 0.73 GB. The full-precision BF16 version is 2.43 GB. The smallest option (Q2_K) is 0.52 GB.
Which GPUs can run Zamba2 1.2B Instruct?
52 consumer GPUs can run Zamba2 1.2B Instruct at Q4_K_M. Top options include AMD Radeon RX 6700 XT, AMD Radeon RX 6800, AMD Radeon RX 6800 XT. 52 GPUs have plenty of headroom for comfortable inference.
Which devices can run Zamba2 1.2B Instruct?
163 devices with unified memory can run Zamba2 1.2B Instruct at Q4_K_M (1.7 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.