Granite 3.1 3B A800m Instruct — Hardware Requirements & GPU Compatibility
ChatFunctions
Granite 3.1 3B A800m Instruct is a 3.3B-parameter open language model from IBM in the Granite family. It supports a context window of up to 131,072 tokens. At Q4_K_M it needs about 2.41 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 · 64K)Best on 24 GB (Q4_K_M · ~364 tok/s · 128K)Macs from 8 GB (Q4_K_M · ~65 tok/s)
Apache 2.0
Reference GPU · RTX 4090 24 GB
Estimates, ±20% · how we estimate
Run Granite 3.1 3B A800m 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/bartowski/granite-3.1-3b-a800m-instruct-GGUF:Q4_K_MQ4_K_M fits fully on the RTX 4090 · 20.6 GB free · 16K context
Download: granite-3.1-3b-a800m-instruct-Q4_K_M.gguf2.02 GBbartowski
Ways to runOther ways
each opens step-by-stepRun in cloud
Runs on RTX 3060 12GB (8 GB headroom) · Q4_K_M
- Generation speed
- ~221 tok/s
- generation speed
- Cost per 1M output tokens
- $0.077
- per 1M output tokens
How Much VRAM Does Granite 3.1 3B A800m Instruct Need?
Select a quantization level; the listed hardware adheres to it.
| Quantization | VRAM | +32K | On RTX 4090 (reference) | |
|---|---|---|---|---|
| 1.8 GB | 3.9 GB | 2-bit · K-quant — 2-bit quantization with K-quant improvements | Fits: 128K context · ~398 tok/s | |
| 1.9 GB | 3.9 GB | 3-bit small — 3-bit small quantization | Fits: 128K context · ~396 tok/s | |
| 2.0 GB | 4.1 GB | 3-bit medium — 3-bit medium quantization | Fits: 128K context · ~386 tok/s | |
| 2.1 GB | 4.1 GB | 4-bit · legacy — 4-bit legacy quantization | Fits: 128K context · ~383 tok/s | |
| 2.4 GB | 4.4 GB | 4-bit medium — 4-bit medium quantization — most popular sweet spot | Fits: 128K context · ~364 tok/s | |
| 2.8 GB | 4.8 GB | 5-bit medium — 5-bit medium quantization — good quality/size tradeoff | Fits: 128K context · ~345 tok/s | |
| 3.2 GB | 5.2 GB | 6-bit, very good — 6-bit quantization, very good quality | Fits: 128K context · ~328 tok/s | |
| 3.7 GB | 5.7 GB | 8-bit, near-lossless — 8-bit quantization, near-lossless | Fits: 128K context · ~305 tok/s |
Selected Q4_K_M
4-bit medium quantization3.4GBat 16K context
- Weights
- 2.0 GB
- KV cache
- 1.1 GB
- Overhead
- 0.3 GB
Fits RTX 4090 (reference) · 20.7 GB free · ~364 tok/s
≥50 tok/s needs ≥54 GB/s · RTX 5090 ~432, RTX 3090 Ti ~364
Which Hardware Can Run Granite 3.1 3B A800m Instruct?
Q4_K_M · 2.4 GBGranite 3.1 3B A800m Instruct (Q4_K_M) requires 2.4 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 4+ GB is recommended. Using the full 131K context window can add up to 8.5 GB, bringing total usage to 10.9 GB. 52 GPUs can run it, including NVIDIA GeForce RTX 5090, NVIDIA GeForce RTX 3090 Ti. 163 devices with unified memory can run Granite 3.1 3B A800m Instruct, including NVIDIA DGX Spark, NVIDIA Jetson AGX Thor Developer Kit.
≤ 11 GB
≤ 11 GB tier →fits · 4.7 GB free · up to 64K ctx
16 GPUs · 1 Mac class · 2 device families
RTX 3080~326Simulate output on RTX 3080RTX 2080 Ti~297Simulate output on RTX 2080 TiRTX 3070 Ti~295Simulate output on RTX 3070 TiM3*8 GB~65Simulate output on iMac 24" M3 (8 GB)Orin Nano~87Simulate output on NVIDIA Jetson Orin Nano 8GB (Super)iPhone~50Simulate output on iPhone 1720 GB
fits · 16.6 GB free · up to 128K ctx
1 GPU
RX 7900 XT~129Simulate output on RX 7900 XT32–40 GB
fits · 28.6 GB free · up to 128K ctx
1 GPU · 3 Mac classes · 4 device families
RTX 5090~432Simulate output on RTX 5090M5 Max36 GB~119Simulate output on Mac Studio M5 Max (36 GB)M532 GB~81Simulate output on MacBook Air 13" M5 (32 GB)Orin~151Simulate output on NVIDIA Jetson AGX Orin 32GBSnapdragon X Elite (X1E-80-100)~95Simulate output on Snapdragon X Elite Copilot+ PC
By VRAM tier · typical GPUs and Macs at Q4_K_M
How good is Granite 3.1 3B A800m Instruct?
llmrun Score blends public benchmark boards into one number; 100 = the average open model.
Not enough data yet for an llmrun Score.
A Score needs results on enough independent boards. How the Composite Score is built
Benchmarks
This is the single board showing results for this model; badges show who ran the test.
- MMLU-Pro
TIGER-LabIndependent
Independent25.4%
#126/146 open
Alternatives to Granite 3.1 3B A800m Instruct
Granite family · 28 models →Similar models
6| Model | Score | VRAM | tok/s | Why it is similar |
|---|---|---|---|---|
| D1 3BLiquid AI | — | 2.6 GB | ~285 | same size |
| SmolLM3 3BHugging Face | — | 3.4 GB | ~285 | same size |
| Deepseek Vl2 TinyDeepSeek | — | 9.4 GB | ~152 | same size |
| Llama 3.2 3B InstructMeta | — | 4.1 GB | ~309 | same size |
| AI21 Jamba Reasoning 3BAI21 Labs | — | 2.3 GB | ~291 | same size |
| Sarashina2.2 3B Instruct v0.1sbintuitions | — | 3.7 GB | ~247 | same size |
Specifications
- Publisher
- IBM
- Family
- Granite
- Parameters
- 3.3B
- Active parameters
- 0.9B per token
- Architecture
- GraniteMoeForCausalLM
- Context Length
- 131,072 tokens
- Vocabulary Size
- 49,155
- Release Date
- 2024-12-06
- License
- Apache 2.0
Frequently Asked Questions
How much VRAM does Granite 3.1 3B A800m Instruct need?
Granite 3.1 3B A800m Instruct requires 2.4 GB of VRAM at Q4_K_M, or 7.0 GB at BF16. Full 131K context adds up to 8.5 GB (10.9 GB total).
VRAM = Weights + KV Cache + Overhead
Weights = 3.3B × 4.8 bits ÷ 8 = 2 GB
KV Cache + Overhead ≈ 0.4 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 ≈ 8.9 GB (at full 131K context)
VRAM usage by quantization
What's the best quantization for Granite 3.1 3B A800m Instruct?
For Granite 3.1 3B A800m Instruct, Q4_K_M (2.4 GB) offers the best balance of quality and VRAM usage. Q4_K_L (2.5 GB) provides better quality if you have the VRAM. The smallest option is IQ3_XS at 1.8 GB.
VRAM requirement by quantization
★ Recommended — best balance of quality and VRAM usage.
Can I run Granite 3.1 3B A800m Instruct on a Mac?
Yes — MacBook Air 13" M3 (8 GB) and 145 other Macs can run Granite 3.1 3B A800m Instruct. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 2.4 GB of usable unified memory (RAM minus macOS overhead).
Can I run Granite 3.1 3B A800m Instruct locally?
Yes — Granite 3.1 3B A800m Instruct can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.
How fast is Granite 3.1 3B A800m Instruct?
At Q4_K_M, Granite 3.1 3B A800m Instruct can reach ~153 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~364 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.
tok/s = 1000 ÷ (active GB ÷ (bandwidth GB/s × efficiency) × 1000 + layers × routing ms)
Mixture-of-Experts: just the active experts get read for each token, along with a fixed routing cost per layer.
Example: NVIDIA B200 → 0.6 GB active ÷ (8000 × 0.65) = 0.124 ms, plus 32 layers × 0.055 ms = 1.760 ms, so 1000 ÷ 1.884 ms = ~531 tok/s
Estimated speed at Q4_K_M (2.4 GB)
Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.
What's the download size of Granite 3.1 3B A800m Instruct?
At Q4_K_M, the download is about 1.98 GB. The full-precision BF16 version is 6.60 GB. The smallest option (IQ3_XS) is 1.36 GB.
Which GPUs can run Granite 3.1 3B A800m Instruct?
52 consumer GPUs can run Granite 3.1 3B A800m 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 Granite 3.1 3B A800m Instruct?
163 devices with unified memory can run Granite 3.1 3B A800m Instruct at Q4_K_M (2.4 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.