Will It Run AI

GoogleGoogle

Gemma 2 2B

現行
huggingfaceHuggingFace
382.0Kダウンロード1.4KいいねJun 2024公開日8K トークンコンテキストGemmaライセンス15 入門品質

Gemma 2 2B (2B parameters) requires approximately 4.6 GB of VRAM with Q4_K_M quantization. For the best balance of quality and speed, we recommend hardware with at least 6 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 Gemma 2 2B on your machine.

Run

lms load gemma-2-2b-it && lms server start

Quick specs

Parameters2B
Architecturedense
Context8K tokens
Modalitytext
Min RAM0.8 GB
Rec. RAM1.2 GB (Q4_K_M)
LicenseGemma
FamilyGemma
✓ Chat

About this model

Gemma 2 2B is Google's lightweight model designed for on-device and edge deployment. Delivers strong text generation and reasoning performance at minimal resource cost.

関連モデル

あなたのハードウェア

検出中...

おすすめ

Intel
コスパ最良B
Intel Arc A380 6GB~$139 — 28 tok/s
NVIDIA
総合最良B
RTX 2060 6GB~$349 — 28 tok/s

最適なハードウェア

Gemma 2 2Bのおすすめ

NVIDIA
RTX 4050 Laptop 6GBB
6 GB
NVIDIA
RTX 2060 6GBB
6 GB
Intel
Intel Arc Pro A40 6GBB
6 GB
Intel
Intel Arc A380 6GBB
6 GB
NVIDIA
GTX 1060 6GBB
6 GB

このモデルを実行

Gemma 2 2B on RTX 4050 Laptop 6GBGemma 2 2B on RTX 2060 6GBGemma 2 2B on Intel Arc Pro A40 6GB

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

QuantBitsVRAMQualityFit
Q2_K
2
0.8 GB
Low—
Q3_K_S
3
1.0 GB
Low—
NVFP4
4
1.1 GB
Medium—
Q4_K_M
4
1.2 GB
Medium—
Q5_K_M
5
1.4 GB
High—
Q6_K
6
1.6 GB
High—
Q8_0
8
2.1 GB
Very High—
F16
16
4.1 GB
Maximum—

Quality benchmarks

Gemma 2 2B benchmark scores

Benchmark verified

Coding

SWE-bench Verified—
HumanEval+40.2%
Aider Polyglot—
LiveCodeBench—

Reasoning

MMLU-Pro17.2%
GPQA Diamond3.2%
MATH-50036.6%
ARC Challenge68.4%

General

Chatbot Arena—
IFEval56.7%

Source: official · 2024-06-27

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

Weights1.2 GB
KV Cache1.6 GB
Runtime1.2 GB
Headroom0.6 GB

よくある質問

FAQ — Gemma 2 2B

関連項目

量子化ガイドスコアリングメソドロジーカリキュレーターを開く