Comparisons of Medium Open Source AI Models (40B-150B)

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

Open source AI models with between 40B to 150B parameters. Models are considered open source (also commonly referred to as open weights) where their weights are accessible to download. This allows self-hosting on your own infrastructure and enables customizing the model such as through fine-tuning. Click on any model to see detailed metrics. For more details including relating to our methodology, see our FAQs.

Qwen3.5 122B A10B and

Mistral Medium 3.5 are the highest intelligence Medium open source models, defined as those with 40B-150B parameters, followed by

Qwen3.5 122B A10B &

NVIDIA Nemotron 3 Super.

Highlights

Openness

Artificial Analysis Openness Index · Higher is better

Updated

Intelligence

Artificial Analysis Intelligence Index · Higher is better

Total Parameters

Trainable parameters in billions

Openness

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

Intelligence

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 v2Updated

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

𝜏³-BankingNew

Agentic tool use

Terminal-Bench v2.1New

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-BriefcaseNew

Agentic knowledge work, (Elo-500)/2000

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.

Size

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 vs. Active Parameters

Active parameters at inference time · Artificial Analysis Intelligence Index

Most attractive quadrant

Reasoning models are indicated by a lightbulb icon.

Intelligence vs. Total Parameters

Artificial Analysis Intelligence Index · Size in parameters (billions)

Most attractive quadrant

Alibaba

InclusionAI

LongCat

MBZUAI Institute of Foundation Models

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).

Further details

							Weights		Provider Benchmarks
Qwen3.5 122B A10B (Reasoning)	Alibaba	32	125B 10B active at inference time	262k	$0.7	141	🤗	+2	View
Mistral Medium 3.5	Mistral	30	128B	256k	$1.2	133	🤗		View
Qwen3.5 122B A10B (Non-reasoning)	Alibaba	28	125B 10B active at inference time	262k	$0.7	148	🤗		View
NVIDIA Nemotron 3 Super 120B A12B (Reasoning)	NVIDIA	25	120.6B 12.7B active at inference time	1.00M	$0.3	243	🤗	+2	View
gpt-oss-120b (high)	OpenAI	24	117B 5.1B active at inference time	131k	$0.2	305	🤗	+23	View
HyperNova 60B 2605	Multiverse Computing	22	58.7B 4.8B active at inference time	131k	$0.1	395	🤗		View
Qwen3 Coder Next	Alibaba	21	79.7B 3B active at inference time	256k	$0.4	127	🤗		View
Mistral Small 4 (Reasoning)	Mistral	21	119B 6.5B active at inference time	256k	$0.2	168	🤗		View
Qwen3 Next 80B A3B (Reasoning)	Alibaba	20	80B 3B active at inference time	262k	$1.1	178	🤗	+5	View
Ling 2.6 Flash	InclusionAI	19	107B 7.4B active at inference time	262k	$0.1	195	🤗		View
Devstral 2	Mistral	19	125B	256k	-	30	🤗		View
gpt-oss-120b (low)	OpenAI	18	117B 5.1B active at inference time	131k	$0.2	329	🤗	+19	View
K2 Think V2	MBZUAI Institute of Foundation Models	17	70B	262k	-	-	🤗	-	View
LongCat Flash Lite	LongCat	17	68.5B 3B active at inference time	256k	-	-	🤗		View
INTELLECT-3	Prime Intellect	16	107B 12B active at inference time	131k	-	-	🤗	-	View
Solar Open 100B (Reasoning)	Upstage	15	102B 12B active at inference time	128k	-	-	🤗	-	View
K2-V2 (high)	MBZUAI Institute of Foundation Models	14	70B	512k	-	-	🤗	-	View
Qwen3 Next 80B A3B Instruct	Alibaba	14	80B 3B active at inference time	262k	$0.7	178	🤗	+4	View
K2-V2 (medium)	MBZUAI Institute of Foundation Models	12	70B	512k	-	-	🤗	-	View
Llama Nemotron Super 49B v1.5 (Reasoning)	NVIDIA	12	49B	128k	$0.1	50	🤗		View
Mistral Small 4 (Non-reasoning)	Mistral	12	119B 6.5B active at inference time	256k	$0.2	153	🤗		View
Sarvam 105B (high)	Sarvam	12	106B 10.3B active at inference time	128k	$0.0	118	🤗		View
Llama 4 Scout	Meta	10	109B 17B active at inference time	10.0M	$0.2	104	🤗	+6	View
Hermes 4 - Llama-3.1 70B (Reasoning)	Nous Research	10	70.6B	128k	$0.2	86	🤗		View
Llama Nemotron Super 49B v1.5 (Non-reasoning)	NVIDIA	9	49B	128k	$0.1	50	🤗		View
Llama 3.3 Instruct 70B	Meta	9	70B	128k	$0.6	89	🤗	+18	View
K2-V2 (low)	MBZUAI Institute of Foundation Models	9	70B	512k	-	-	🤗	-	View
Kimi Linear 48B A3B Instruct	Kimi	9	49.1B 3B active at inference time	1.00M	-	-	🤗	-	View
Ring-flash-2.0	InclusionAI	8	103B 6.1B active at inference time	128k	$0.2	-	🤗		View
Command A	Cohere	8	111B	256k	$3.3	67	🤗		View
Llama 3.1 Nemotron Instruct 70B	NVIDIA	8	70B	128k	$1.2	299	🤗		View
Hermes 4 - Llama-3.1 70B (Non-reasoning)	Nous Research	7	70.6B	128k	$0.2	84	🤗		View
Llama 3.2 Instruct 90B (Vision)	Meta	6	90B	128k	$1.4	58	🤗		View
Jamba 1.7 Mini	AI21 Labs	3	52B 12B active at inference time	258k	-	-	🤗	-	View
Apertus 70B Instruct	Swiss AI Initiative	2	70B	65.5k	$1.0	-	🤗		View

Comparisons of Medium Open Source AI Models (40B-150B)

Openness

Intelligence

Total Parameters

Openness

Artificial Analysis Openness Index: Score

Intelligence

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index

Intelligence Evaluations

Intelligence Evaluation Relevance

Artificial Analysis Intelligence Index

Size

Model Size: Total and Active Parameters

Total Parameters

Active Parameters at Inference Time

Intelligence vs. Active Parameters

Artificial Analysis Intelligence Index

Active Parameters at Inference Time

Intelligence vs. Total Parameters

Artificial Analysis Intelligence Index

Total Parameters

Context Window

Context Window

Context Window for RAG

Context Window

Further details