中型开放权重 AI 模型比较（40B-150B）

Name: 开放性
Creator: Artificial Analysis
License: https://artificialanalysis.ai/docs/legal/Terms-of-Use.pdf

参数量介于 40B 和 150B 之间的开放权重 AI 模型。

如果模型权重可供下载，我们便将其视为开放权重模型。这样用户就可以在自己的基础设施上自行托管，并通过微调等方式定制模型。

如需了解包括方法论在内的更多详情，请参阅常见问题。

Qwen3.5 122B A10B 和

Mistral Medium 3.5 是智能得分最高的中型开放权重模型，定义为参数量介于 40B 和 150B 之间的模型，其次是

Qwen3.5 122B A10B 和

NVIDIA Nemotron 3 Super。

亮点

开放性

Artificial Analysis Openness Index · Higher is better

智能

Artificial Analysis Intelligence Index · Higher is better

开放性

Artificial Analysis Openness Index: Score

Openness Index assesses model openness on a 0 to 100 normalized scale (higher is more open)

Reasoning models are indicated by a lightbulb icon

智能

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v4.1 incorporates 9 evaluations: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR

Estimate (independent evaluation forthcoming)

Reasoning models are indicated by a lightbulb icon

Artificial Analysis Intelligence Index v4.1 includes: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.

Intelligence Evaluations

Intelligence evaluations measured independently by Artificial Analysis · Higher is better

GDPval-AA v2

Agentic real-world work tasks, (Elo-500)/2000

𝜏³-Banking

Agentic tool use

Terminal-Bench v2.1

Agentic coding & terminal use

SciCode

Coding

Humanity's Last Exam

Reasoning & knowledge

GPQA Diamond

Scientific reasoning

CritPt

Physics reasoning

AA-Omniscience Accuracy

Knowledge

AA-Omniscience Non-Hallucination Rate

1 - hallucination rate

AA-LCR

Long context reasoning

AA-Briefcase

Agentic knowledge work, Elo

AutomationBench-AA

Agentic SaaS workflows

Harvey LAB-AA

Legal agentic work, criterion pass rate

EnterpriseOps-Gym-AANew

Agentic business operations

IFBench

Instruction following

APEX-Agents-AA

Long-horizon agentic tasks

ITBench-AA

Kubernetes incident root-cause analysis

MMMU-Pro

Visual reasoning

Reasoning models are indicated by a lightbulb icon

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

规模

Model Size: Total and Active Parameters

Comparison between total model parameters and parameters active during inference

Reasoning models are indicated by a lightbulb icon

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 vs. Active Parameters

Artificial Analysis Intelligence Index · Active parameters at inference time

Most attractive quadrant

Pareto line

Reasoning models are indicated by a lightbulb icon

Intelligence Index vs. Total Parameters

Artificial Analysis Intelligence Index · Size in parameters (billions)

Most attractive quadrant

Pareto line

Reasoning models are indicated by a lightbulb icon

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.

上下文窗口

Context Window

Context window: tokens limit · Higher is better

Reasoning models are indicated by a lightbulb icon

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).

							权重		服务商基准测试
Qwen3.5 122B A10B (Reasoning)	Alibaba	32	125B 推理时启用 10B 个参数	262k	$0.7	125	🤗	+2	查看
Mistral Medium 3.5	Mistral	30	128B	256k	$1.2	134	🤗		查看
Qwen3.5 122B A10B (Non-reasoning)	Alibaba	28	125B 推理时启用 10B 个参数	262k	$0.7	140	🤗		查看
NVIDIA Nemotron 3 Super 120B A12B (Reasoning)	NVIDIA	25	120.6B 推理时启用 12.7B 个参数	1M	$0.3	143	🤗		查看
gpt-oss-120b (high)	OpenAI	24	117B 推理时启用 5.1B 个参数	131k	$0.2	185	🤗	+18	查看
Qwen3 Coder Next	Alibaba	21	79.7B 推理时启用 3B 个参数	256k	$0.4	133	🤗		查看
Mistral Small 4 (Reasoning)	Mistral	20	119B 推理时启用 6.5B 个参数	256k	$0.2	158	🤗		查看
Devstral 2	Mistral	19	125B	256k	-	43	🤗		查看
HyperNova 60B 2605	Multiverse Computing	18	58.7B 推理时启用 4.8B 个参数	131k	$0.1	349	🤗		查看
K2 Think V2	MBZUAI Institute of Foundation Models	17	70B	262k	-	-	🤗	-	查看
LongCat Flash Lite	LongCat	17	68.5B 推理时启用 3B 个参数	256k	-	-	🤗	-	查看
Qwen3 Next 80B A3B (Reasoning)	Alibaba	17	80B 推理时启用 3B 个参数	262k	$1.1	205	🤗	+2	查看

中型开放权重 AI 模型比较（40B-150B）

开放性

智能

总参数量

开放性

Artificial Analysis Openness Index: Score

智能

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index

Intelligence Evaluations

Intelligence Evaluation Relevance

Artificial Analysis Intelligence Index

规模

Model Size: Total and Active Parameters

Total Parameters

Active Parameters at Inference Time

Intelligence Index vs. Active Parameters

Artificial Analysis Intelligence Index

Active Parameters at Inference Time

Intelligence Index vs. Total Parameters

Artificial Analysis Intelligence Index

Total Parameters

上下文窗口

Context Window

Context Window for RAG

Context Window

更多详情