Only with CPU offload

Granite 4.1 30B at Q4_K_M needs 18.9 GB but only 8.8 GB is addressable, so about 62% of the layers would stream from system RAM at roughly 60 GB/s. Expect around 3.3 tokens per second — usable for batch work, painful for chat.

38% on GPU 8K context Q4_K_M · 16.3 GB Apache 2.0 Released 29 Apr 2026

The largest Granite. Dense 29B at Q4 is a comfortable 24 GB fit.

Hardware requirements What hardware do I need to run Granite 4.1 30B? →

At Q4_K_M with an 8K context: a 24 GB GPU (RTX 4090) or a 32 GB Mac (M4). Minimum, recommended, high-quality and long-context tiers, with all 114 devices checked at every quantisation.

Fits instead: Granite 4.1 8B (6.8 GB) · Granite 4.1 3B (3.1 GB)

The VRAM budget

weights 16.3 GB
Weights 16.3 GB KV cache @ 8K 2.00 GB Runtime overhead 0.6 GB Over budget 10.1 GB past 8.8 GB

Quantisation ladder

QuantWeightsTotal @ 8KMax contextTok/sQualityFit
Q8_0 28.6 GB 31.2 GB — ~1.6 −0.1% ppl 22.4 GB over
Q6_K 22.1 GB 24.7 GB — ~2.2 −0.4% ppl 15.9 GB over
Q5_K_M 19.1 GB 21.7 GB — ~2.6 −0.8% ppl 12.9 GB over
Q4_K_M 16.3 GB 18.9 GB — ~3.3 −1.9% ppl 10.1 GB over
Q3_K_M 13.2 GB 15.8 GB — ~4.6 −5.4% ppl 7.0 GB over
Q2_K 11.3 GB 13.9 GB — ~6.0 −15% ppl 5.1 GB over

Quality is the published perplexity delta against f16 weights. Max context assumes an f16 KV cache; q8_0 roughly doubles it.

How to run it

terminal
$ llama-server \
    -hf ibm-granite/granite-4.1-30b:Q4_K_M \
    -c 8192 -ngl 24

The engine underneath most of the others. Every knob is exposed. More on llama.cpp.

01Download is 16.3 GB. Keep it on an SSD — a first load off a spinning disk takes minutes.
02Close anything else holding VRAM. A browser with hardware acceleration can sit on 1–2 GB.
03If it falls back to CPU silently, drop the context first, then step down a quantisation.
See all models for this rig Compare with another model

Related models on an Arc B570

ModelTotalTok/s est.
Granite 4.1 8B
8.79B · same family
Q4_K_M · 6.8 GB 47 Runs great
Granite 4.1 3B
3.4B · same family
Q4_K_M · 3.1 GB 120 Runs great
Muse Glimmer 30B
29.8B · similar size
Q4_K_M · 17.5 GB ~3.6 CPU offload
Qwen3.8 27B
27.8B · similar size
Q4_K_M · 16.7 GB ~4.0 CPU offload
Qwen3.6 27B
27.8B · similar size
Q4_K_M · 16.7 GB ~4.0 CPU offload
Gemma 3 27B
27.4B · similar size
Q4_K_M · 17.0 GB ~3.9 CPU offload
Qwen3 30B-A3B
30.5B MoE · similar size
Q4_K_M · 18.5 GB ~12 CPU offload

Every model against the Arc B570 →

Granite 4.1 30B on other hardware

DeviceTotalTok/s est.
GeForce RTX 4060 Laptop
7.0 GB usable · 256 GB/s
18.9 GB ~2.8 CPU offload
GeForce RTX 3060 12 GB
10.6 GB usable · 360 GB/s
18.9 GB ~3.8 CPU offload
M4 · 16 GB
10.7 GB usable · 120 GB/s
18.9 GB — Won't fit
GeForce RTX 4090
22.4 GB usable · 1008 GB/s
18.9 GB 38 Runs great
Radeon RX 7900 XTX
22.4 GB usable · 960 GB/s
18.9 GB 36 Runs great
M4 · 32 GB
24.0 GB usable · 120 GB/s
18.9 GB 4.1 Runs great
M3 Max · 64 GB
48.0 GB usable · 400 GB/s
18.9 GB 14 Runs great

All 114 devices, by tier →