中規模オープンウェイトAIモデルの比較（40B～150B）

Name: オープン性
Creator: Artificial Analysis
License: https://artificialanalysis.ai/docs/legal/Terms-of-Use.pdf

パラメーター数が40B～150BのオープンウェイトAIモデルです。

ウェイトをダウンロードできるモデルをオープンウェイト（一般にオープンソースとも呼ばれます）とみなします。独自のインフラストラクチャでセルフホストでき、ファインチューニングなどによるモデルのカスタマイズも可能です。

方法論などの詳細は、よくある質問をご覧ください。

Nemotron 3.5 Super VLと

Ling-3.0-flash-VLはパラメーター数40B～150Bと定義される中規模オープンウェイトモデルの中で知能が最も高く、

Ling-3.0-flash-Finと

Nemotron 3.5 Bolt 74B-A9Bが続きます。

オープン性

Artificial Analysis Openness Index · Higher is better

知能

Artificial Analysis Intelligence Index · Higher is better

オープン性

Artificial Analysis Openness Index：スコア

Openness Indexはモデルのオープン性を0～100の正規化尺度で評価します（高いほどオープン）

知能

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v4.3.2 incorporates 10 evaluations: AA-Briefcase v1.1, GDPval-AA v2.1, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, AA-LCR v1.1

Estimate (independent evaluation forthcoming)

Artificial Analysis Intelligence Index v4.3.2 includes: AA-Briefcase v1.1, GDPval-AA v2.1, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, AA-LCR v1.1. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.

Evaluations

Evaluations measured independently by Artificial Analysis · AA-PersonalityBench scores are percentile ranks from 1 to 100

AA-Briefcase v1.1

Agentic knowledge work, Elo

AA-Omniscience Accuracy

Knowledge

AA-Omniscience Non-Hallucination Rate

1 - hallucination rate

AutomationBench-AA

Agentic SaaS workflows

GDPval-AA v2.1

Agentic real-world work tasks, Elo

Humanity's Last Exam

Reasoning & knowledge

GDP.pdf

Professional document reasoning, All-pass

Terminal-Bench 4.0

Agentic coding & terminal use

AA-LCR v1.1

Long context reasoning

SciCodeUnder review

Coding

CritPt

Physics reasoning

Terminal-Bench-Science 0.1New

Agentic scientific research workflows in a terminal

FrontierCode 1.1New

Agentic software engineering in open-source repositories

AA-AnalystAgent

Quantitative analysis on spreadsheets & documents

Harvey LAB-AA v1.1Updated

Legal agentic work, Hallucination-Gated All-Pass Rate

No data available

EnterpriseOps-Gym-AA

Agentic business operations

AA-3DNew

Agentic interactive 3D experiences, Elo

No data available

ITBench-AA

Kubernetes incident root-cause analysis

MLCR-AAUpdated

Medical long context reasoning

MMMU-Pro

Visual reasoning

AA-PersonalityBench: Expressiveness ScoreNew

Rich, varied, well-crafted language

AA-PersonalityBench: Organization ScoreNew

Well-structured, scannable, answer-first

AA-PersonalityBench: Transparency ScoreNew

Truthful, acknowledges uncertainty, non-sycophantic

AA-PersonalityBench: Slop ScoreNew

Use of filler and generic language

While model intelligence generally translates across use cases, specific evaluations may be more relevant for certain use cases.

規模

モデルサイズ：総パラメータ数とアクティブパラメータ数

Comparison between total model parameters and parameters active during inference (billions)

The total number of trainable weights and biases in the model, expressed in billions. These parameters are learned during training and determine the model's ability to process and generate responses.

The number of parameters actually executed during each inference forward pass, expressed in billions. For Mixture of Experts (MoE) models, a routing mechanism selects a subset of experts per token, resulting in fewer active than total parameters. Dense models use all parameters, so active equals total.

Intelligence Indexと有効パラメーター数

Artificial Analysis Intelligence Index · Active parameters at inference time (billions)

Most attractive quadrant

Pareto line

Intelligence Index と総パラメータ数

Artificial Analysis Intelligence Index · Size in parameters (billions)

Most attractive quadrant

Pareto line

The total number of trainable weights and biases in the model, expressed in billions. These parameters are learned during training and determine the model's ability to process and generate responses.

コンテキストウィンドウ

Context window: tokens limit · Higher is better

Larger context windows are relevant to RAG (Retrieval Augmented Generation) LLM workflows which typically involve reasoning and information retrieval of large amounts of data.

Maximum number of combined input & output tokens. Output tokens commonly have a significantly lower limit (varied by model).

詳細

							ウェイト		プロバイダーのベンチマーク
Nemotron 3.5 Super VL	NVIDIA	26	120B 推論時に12Bが有効	1M	$0.1	446	🤗	+5	表示
Ling-3.0-flash-VL	InclusionAI	25	124B 推論時に5.5Bが有効	262k	-	-	🤗	-	表示
Ling-3.0-flash-Fin	InclusionAI	23	124B 推論時に5.1Bが有効	262k	-	-	🤗	-	表示
Nemotron 3.5 Bolt 74B-A9B	NVIDIA	21	74B 推論時に9Bが有効	262k	-	-	🤗	-	表示
Qwen3.5 122B A10B (Non-reasoning)	Alibaba	18	125B 推論時に10Bが有効	262k	$0.7	143	🤗		表示
Qwen3.5 122B A10B (Reasoning)	Alibaba	16	125B 推論時に10Bが有効	262k	$0.7	128	🤗		表示
Mistral Medium 3.5	Mistral	14	128B	256k	$1.2	162	🤗		表示
HyperNova 60B 2605 (High, Based on gpt-oss-120b)	Multiverse Computing	12	58.7B 推論時に4.8Bが有効	131k	-	-	🤗	-	表示
gpt-oss-120b (High)	OpenAI	12	117B 推論時に5.1Bが有効	131k	$0.2	233	🤗	+14	表示
K2 Think V2	Institute of Foundation Models	11	70B	262k	-	-	🤗	-	表示
LongCat Flash Lite	LongCat	11	68.5B 推論時に3Bが有効	256k	-	-	🤗	-	表示
Mistral Small 4 (Reasoning)	Mistral	11	119B 推論時に6.5Bが有効	256k	$0.1	178	🤗		表示

中規模オープンウェイトAIモデルの比較（40B～150B）

オープン性

知能

総パラメーター数

オープン性

Artificial Analysis Openness Index：スコア

知能

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index

Evaluations

Intelligence Evaluation Relevance

Artificial Analysis Intelligence Index

規模

モデルサイズ：総パラメータ数とアクティブパラメータ数

Total Parameters

Active Parameters at Inference Time

Intelligence Indexと有効パラメーター数

Artificial Analysis Intelligence Index

Active Parameters at Inference Time

Intelligence Index と総パラメータ数

Artificial Analysis Intelligence Index

Total Parameters

コンテキストウィンドウ

コンテキストウィンドウ

Context Window for RAG

Context Window

詳細