IBM

Granite 4.1 30B

現行

HuggingFace

Ollama

32.1Kダウンロード126いいねApr 2026公開日131K トークンコンテキストApache 2.0ライセンス76 優秀品質

Granite 4.1 30B (30B parameters) requires approximately 24.0 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 28 GB of VRAM.

はじめに

— コピー&ペーストでローカル実行

Copy-paste commands to run Granite 4.1 30B on your machine.

Run

ollama run granite4.1:30b

Quick specs

Parameters30B

Architecturedense

Context131K tokens

Modalitytext

Min RAM11.7 GB

Rec. RAM18.3 GB (Q4_K_M)

LicenseApache 2.0

FamilyGranite

✓ Code✓ Chat✓ Reasoning✓ RAG

About this model

Granite 4.1 30B is IBM's serious local workstation dense decoder-only model, trained on roughly 15T tokens with 128K context. Q4 fits on a 24 GB GPU with room for KV cache, making it a strong enterprise-friendly assistant for RAG and coding. Apache 2.0 licensed.

•Dense 30B runs on 24 GB GPUs (RTX 4090/3090) at Q4
•Apache 2.0 license for commercial RAG, coding, and assistant use
•128K context for long-document and agent workflows

Granite 4.1 30Bのおすすめ

NVIDIA A100 40GBS

40 GB

MacBook Pro M4 Max 48GBS

48 GB

RTX PRO 5000 Blackwell 48GBS

48 GB

RTX 5090 32GBS

32 GB

MacBook Pro M4 Max 64GBS

64 GB

このモデルを実行

Granite 4.1 30B on NVIDIA A100 40GB Granite 4.1 30B on MacBook Pro M4 Max 48GB Granite 4.1 30B on RTX PRO 5000 Blackwell 48GB

量子化オプション

量子化レベル別VRAM推定値

No hardware detected — fit column shows raw VRAM estimates

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	11.7 GB	Low	—
Q3_K_S	3	14.7 GB	Low	—
NVFP4	4	16.8 GB	Medium	—
Q4_K_M	4	18.3 GB	Medium	—
Q5_K_M	5	21.6 GB	High	—
Q6_K	6	24.6 GB	High	—
Q8_0	8	32.1 GB	Very High	—
F16	16	61.5 GB	Maximum	—

ハードウェア互換性

全ハードウェアの適合度推定

カリキュレーターを開く

Computing compatibility...

メモリ内訳

Reference: RTX 2060 6GB

Weights18.3 GB

KV Cache3.9 GB

Runtime1.2 GB

Headroom0.6 GB

よくある質問

Granite 4.1 30B

Granite 4.1 30Bのおすすめ

量子化レベル別VRAM推定値

全ハードウェアの適合度推定

Reference: RTX 2060 6GB

FAQ — Granite 4.1 30B