Comparison of Models: Intelligence, Performance & Price Analysis
Comparison and analysis of AI models across key performance metrics including quality, price, output speed, latency, context window & others. Click on any model to see detailed metrics. For more details including relating to our methodology, see our FAQs.
Model Comparison Summary
Gemini 3 Pro Preview and 
GPT-5.1 (high) are the highest intelligence models, followed by GPT-5 Codex (high) & GPT-5 (high).Output Speed (tokens/s): Gemini 2.5 Flash-Lite (Sep) (475 t/s) and 
Granite 3.3 8B (423 t/s) are the fastest models, followed by Gemini 2.5 Flash & Granite 4.0 H Small.Latency (seconds):
DeepSeek-OCR (0.19s) and 
Apriel-v1.5-15B-Thinker (0.25s) are the lowest latency models, followed by Gemini 2.0 Flash-Lite (Feb) & 
Mistral Small (Feb).Price ($ per M tokens): Gemma 3n E4B ($0.03) and Ministral 3B ($0.04) are the cheapest models, followed by Gemma 2 9B & DeepSeek-OCR.Context Window:
Llama 4 Scout (10m) and 
MiniMax-Text-01 (4m) are the largest context window models, followed by 
Grok 4 Fast & Grok 4.1 Fast.Highlights
Intelligence
Artificial Analysis Intelligence Index
Combination metric covering multiple dimensions of intelligence - the simplest way to compare how smart models are. Version 3.0 was released in September 2025 and includes: MMLU-Pro, GPQA Diamond, Humanity's Last Exam, LiveCodeBench, SciCode, AIME 2025, IFBench, AA-LCR, Terminal-Bench Hard, 𝜏²-Bench Telecom. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.
{"@context":"https://schema.org","@type":"Dataset","name":"Artificial Analysis Intelligence Index","creator":{"@type":"Organization","name":"Artificial Analysis","url":"https://artificialanalysis.ai"},"description":"Artificial Analysis Intelligence Index v3.0 incorporates 10 evaluations: MMLU-Pro, GPQA Diamond, Humanity's Last Exam, LiveCodeBench, SciCode, AIME 2025, IFBench, AA-LCR, Terminal-Bench Hard, 𝜏²-Bench Telecom","measurementTechnique":"Independent test run by Artificial Analysis on dedicated hardware.","spatialCoverage":"Worldwide","keywords":["analytics","llm","AI","benchmark","model","gpt","claude"],"license":"https://creativecommons.org/licenses/by/4.0/","isAccessibleForFree":true,"citation":"Artificial Analysis (2025). LLM benchmarks dataset. https://artificialanalysis.ai","data":""}
Artificial Analysis Coding Index
Represents the average of coding evaluations in the Artificial Analysis Intelligence Index. Currently includes: LiveCodeBench, SciCode, Terminal-Bench Hard. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.
{"@context":"https://schema.org","@type":"Dataset","name":"Artificial Analysis Coding Index","creator":{"@type":"Organization","name":"Artificial Analysis","url":"https://artificialanalysis.ai"},"description":"Represents the average of coding benchmarks in the Artificial Analysis Intelligence Index (LiveCodeBench, SciCode, Terminal-Bench Hard)","measurementTechnique":"Independent test run by Artificial Analysis on dedicated hardware.","spatialCoverage":"Worldwide","keywords":["analytics","llm","AI","benchmark","model","gpt","claude"],"license":"https://creativecommons.org/licenses/by/4.0/","isAccessibleForFree":true,"citation":"Artificial Analysis (2025). LLM benchmarks dataset. https://artificialanalysis.ai","data":""}
Artificial Analysis Agentic Index
Represents the average of agentic capabilities benchmarks in the Artificial Analysis Intelligence Index (Terminal-Bench Hard, 𝜏²-Bench Telecom).
{"@context":"https://schema.org","@type":"Dataset","name":"Artificial Analysis Agentic Index","creator":{"@type":"Organization","name":"Artificial Analysis","url":"https://artificialanalysis.ai"},"description":"Represents the average of agentic capabilities benchmarks in the Artificial Analysis Intelligence Index (Terminal-Bench Hard, 𝜏²-Bench Telecom)","measurementTechnique":"Independent test run by Artificial Analysis on dedicated hardware.","spatialCoverage":"Worldwide","keywords":["analytics","llm","AI","benchmark","model","gpt","claude"],"license":"https://creativecommons.org/licenses/by/4.0/","isAccessibleForFree":true,"citation":"Artificial Analysis (2025). LLM benchmarks dataset. https://artificialanalysis.ai","data":""}
Artificial Analysis Intelligence Index by Open Weights vs Proprietary
Combination metric covering multiple dimensions of intelligence - the simplest way to compare how smart models are. Version 3.0 was released in September 2025 and includes: MMLU-Pro, GPQA Diamond, Humanity's Last Exam, LiveCodeBench, SciCode, AIME 2025, IFBench, AA-LCR, Terminal-Bench Hard, 𝜏²-Bench Telecom. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.
Indicates whether the model weights are available. Models are labelled as 'Commercial Use Restricted' if the weights are available but commercial use is limited (typically requires obtaining a paid license).
{"@context":"https://schema.org","@type":"Dataset","name":"Artificial Analysis Intelligence Index by Open Weights vs Proprietary","creator":{"@type":"Organization","name":"Artificial Analysis","url":"https://artificialanalysis.ai"},"description":"Artificial Analysis Intelligence Index v3.0 incorporates 10 evaluations: MMLU-Pro, GPQA Diamond, Humanity's Last Exam, LiveCodeBench, SciCode, AIME 2025, IFBench, AA-LCR, Terminal-Bench Hard, 𝜏²-Bench Telecom","measurementTechnique":"Independent test run by Artificial Analysis on dedicated hardware.","spatialCoverage":"Worldwide","keywords":["analytics","llm","AI","benchmark","model","gpt","claude"],"license":"https://creativecommons.org/licenses/by/4.0/","isAccessibleForFree":true,"citation":"Artificial Analysis (2025). LLM benchmarks dataset. https://artificialanalysis.ai","data":""}
Artificial Analysis Intelligence Index by Model Type
Combination metric covering multiple dimensions of intelligence - the simplest way to compare how smart models are. Version 3.0 was released in September 2025 and includes: MMLU-Pro, GPQA Diamond, Humanity's Last Exam, LiveCodeBench, SciCode, AIME 2025, IFBench, AA-LCR, Terminal-Bench Hard, 𝜏²-Bench Telecom. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.
{"@context":"https://schema.org","@type":"Dataset","name":"Artificial Analysis Intelligence Index by Model Type","creator":{"@type":"Organization","name":"Artificial Analysis","url":"https://artificialanalysis.ai"},"description":"Artificial Analysis Intelligence Index v3.0 incorporates 10 evaluations: MMLU-Pro, GPQA Diamond, Humanity's Last Exam, LiveCodeBench, SciCode, AIME 2025, IFBench, AA-LCR, Terminal-Bench Hard, 𝜏²-Bench Telecom","measurementTechnique":"Independent test run by Artificial Analysis on dedicated hardware.","spatialCoverage":"Worldwide","keywords":["analytics","llm","AI","benchmark","model","gpt","claude"],"license":"https://creativecommons.org/licenses/by/4.0/","isAccessibleForFree":true,"citation":"Artificial Analysis (2025). LLM benchmarks dataset. https://artificialanalysis.ai","data":""}
Intelligence Evaluations
While model intelligence generally translates across use cases, specific evaluations may be more relevant for certain use cases.
Combination metric covering multiple dimensions of intelligence - the simplest way to compare how smart models are. Version 3.0 was released in September 2025 and includes: MMLU-Pro, GPQA Diamond, Humanity's Last Exam, LiveCodeBench, SciCode, AIME 2025, IFBench, AA-LCR, Terminal-Bench Hard, 𝜏²-Bench Telecom. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.
Artificial Analysis Omniscience
AA-Omniscience Index
AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct.
{"@context":"https://schema.org","@type":"Dataset","name":"AA-Omniscience Index","creator":{"@type":"Organization","name":"Artificial Analysis","url":"https://artificialanalysis.ai"},"description":"AA-Omniscience Index (higher is better) measures knowledge reliability and hallucination. It rewards correct answers, penalizes hallucinations, and has no penalty for refusing to answer. Scores range from -100 to 100, where 0 means as many correct as incorrect answers, and negative scores mean more incorrect than correct.","measurementTechnique":"Independent test run by Artificial Analysis on dedicated hardware.","spatialCoverage":"Worldwide","keywords":["analytics","llm","AI","benchmark","model","gpt","claude"],"license":"https://creativecommons.org/licenses/by/4.0/","isAccessibleForFree":true,"citation":"Artificial Analysis (2025). LLM benchmarks dataset. https://artificialanalysis.ai","data":"modelName,omniscienceIndex,detailsUrl,isLabClaimedValue\nGemini 3 Pro Preview,12.867,/models/gemini-3-pro/providers,false\nClaude 4.1 Opus,4.933,/models/claude-4-1-opus-thinking/providers,false\nGPT-5.1 (high),2.2,/models/gpt-5-1/providers,false\nGrok 4,0.95,/models/grok-4/providers,false\nClaude 4.5 Sonnet,-2.083,/models/claude-4-5-sonnet-thinking/providers,false\nClaude 4.5 Haiku,-5.667,/models/claude-4-5-haiku-reasoning/providers,false\nGPT-5 (high),-11.1,/models/gpt-5/providers,false\nGPT-5 (low),-12.933,/models/gpt-5-low/providers,false\nGPT-5 (medium),-13.733,/models/gpt-5-medium/providers,false\nGemini 2.5 Pro,-17.95,/models/gemini-2-5-pro/providers,false\nLlama 3.1 405B,-18.167,/models/llama-3-1-instruct-405b/providers,false\nGPT-5 mini (high),-19.617,/models/gpt-5-mini/providers,false\nKimi K2 Thinking,-23.417,/models/kimi-k2-thinking/providers,false\nDeepSeek V3.1 Terminus,-26.7,/models/deepseek-v3-1-terminus-reasoning/providers,false\nMagistral Medium 1.2,-27.633,/models/magistral-medium-2509/providers,false\nKimi K2 0905,-28.35,/models/kimi-k2-0905/providers,false\nDeepSeek R1 0528,-29.667,/models/deepseek-r1/providers,false\nGrok 4 Fast,-30.483,/models/grok-4-fast-reasoning/providers,false\nGrok 4.1 Fast,-31.433,/models/grok-4-1-fast-reasoning/providers,false\nDeepSeek V3.2 Exp,-31.9,/models/deepseek-v3-2-reasoning/providers,false\nGPT-5.1,-36.583,/models/gpt-5-1-non-reasoning/providers,false\nGPT-5 (minimal),-36.667,/models/gpt-5-minimal/providers,false\nGemini 2.5 Flash (Sep),-37.5,/models/gemini-2-5-flash-preview-09-2025-reasoning/providers,false\nGPT-4.1,-42.133,/models/gpt-4-1/providers,false\nNVIDIA Nemotron Nano 9B V2,-43.217,/models/nvidia-nemotron-nano-9b-v2-reasoning/providers,false\nLlama 4 Maverick,-43.467,/models/llama-4-maverick/providers,false\nGLM-4.6,-43.883,/models/glm-4-6-reasoning/providers,false\nQwen3 235B 2507,-45.383,/models/qwen3-235b-a22b-instruct-2507/providers,false\nLlama Nemotron Super 49B v1.5,-47.467,/models/llama-nemotron-super-49b-v1-5-reasoning/providers,false\nQwen3 235B A22B 2507,-47.7,/models/qwen3-235b-a22b-instruct-2507-reasoning/providers,false\nMiniMax-M2,-49.533,/models/minimax-m2/providers,false\nCommand A,-50.4,/models/command-a/providers,false\ngpt-oss-120B (high),-51.933,/models/gpt-oss-120b/providers,false\nQwen3 Next 80B A3B,-52.783,/models/qwen3-next-80b-a3b-reasoning/providers,false\ngpt-oss-120B (low),-55.933,/models/gpt-oss-120b-low/providers,false\ngpt-oss-20B (low),-60.6,/models/gpt-oss-20b-low/providers,false\nEXAONE 4.0 32B,-61.417,/models/exaone-4-0-32b-reasoning/providers,false\nGranite 4.0 H Small,-62.067,/models/granite-4-0-h-small/providers,false\ngpt-oss-20B (high),-64.9,/models/gpt-oss-20b/providers,false\nQwen3 8B,-66.117,/models/qwen3-8b-instruct-reasoning/providers,false\nQwen3 8B,-75.4,/models/qwen3-8b-instruct/providers,false"}
AA-Omniscience Accuracy
AA-Omniscience Accuracy (higher is better) measures the proportion of correctly answered questions out of all questions, regardless of whether the model chooses to answer
{"@context":"https://schema.org","@type":"Dataset","name":"AA-Omniscience Accuracy","creator":{"@type":"Organization","name":"Artificial Analysis","url":"https://artificialanalysis.ai"},"description":"AA-Omniscience Accuracy (higher is better) measures the proportion of correctly answered questions out of all questions, regardless of whether the model chooses to answer","measurementTechnique":"Independent test run by Artificial Analysis on dedicated hardware.","spatialCoverage":"Worldwide","keywords":["analytics","llm","AI","benchmark","model","gpt","claude"],"license":"https://creativecommons.org/licenses/by/4.0/","isAccessibleForFree":true,"citation":"Artificial Analysis (2025). LLM benchmarks dataset. https://artificialanalysis.ai","data":"modelName,omniscienceAccuracy,detailsUrl,isLabClaimedValue\nGemini 3 Pro Preview,0.5365,/models/gemini-3-pro/providers,false\nGrok 4,0.39466666666666667,/models/grok-4/providers,false\nGPT-5 (high),0.38616666666666666,/models/gpt-5/providers,false\nGemini 2.5 Pro,0.37483333333333335,/models/gemini-2-5-pro/providers,false\nGPT-5 (medium),0.37383333333333335,/models/gpt-5-medium/providers,false\nGPT-5 (low),0.3635,/models/gpt-5-low/providers,false\nClaude 4.1 Opus,0.35933333333333334,/models/claude-4-1-opus-thinking/providers,false\nGPT-5.1 (high),0.353,/models/gpt-5-1/providers,false\nClaude 4.5 Sonnet,0.309,/models/claude-4-5-sonnet-thinking/providers,false\nDeepSeek R1 0528,0.29283333333333333,/models/deepseek-r1/providers,false\nKimi K2 Thinking,0.29233333333333333,/models/kimi-k2-thinking/providers,false\nGPT-5.1,0.278,/models/gpt-5-1-non-reasoning/providers,false\nGPT-5 (minimal),0.27216666666666667,/models/gpt-5-minimal/providers,false\nDeepSeek V3.1 Terminus,0.27166666666666667,/models/deepseek-v3-1-terminus-reasoning/providers,false\nGemini 2.5 Flash (Sep),0.2698333333333333,/models/gemini-2-5-flash-preview-09-2025-reasoning/providers,false\nDeepSeek V3.2 Exp,0.26966666666666667,/models/deepseek-v3-2-reasoning/providers,false\nGPT-4.1,0.2608333333333333,/models/gpt-4-1/providers,false\nGLM-4.6,0.25483333333333336,/models/glm-4-6-reasoning/providers,false\nKimi K2 0905,0.24033333333333334,/models/kimi-k2-0905/providers,false\nLlama 4 Maverick,0.23433333333333334,/models/llama-4-maverick/providers,false\nGrok 4.1 Fast,0.23433333333333334,/models/grok-4-1-fast-reasoning/providers,false\nGPT-5 mini (high),0.22966666666666666,/models/gpt-5-mini/providers,false\nQwen3 235B A22B 2507,0.22116666666666668,/models/qwen3-235b-a22b-instruct-2507-reasoning/providers,false\nGrok 4 Fast,0.21933333333333332,/models/grok-4-fast-reasoning/providers,false\nLlama 3.1 405B,0.21733333333333332,/models/llama-3-1-instruct-405b/providers,false\nMiniMax-M2,0.20833333333333334,/models/minimax-m2/providers,false\nMagistral Medium 1.2,0.20083333333333334,/models/magistral-medium-2509/providers,false\ngpt-oss-120B (high),0.19983333333333334,/models/gpt-oss-120b/providers,false\nQwen3 Next 80B A3B,0.18166666666666667,/models/qwen3-next-80b-a3b-reasoning/providers,false\ngpt-oss-120B (low),0.1815,/models/gpt-oss-120b-low/providers,false\nQwen3 235B 2507,0.1755,/models/qwen3-235b-a22b-instruct-2507/providers,false\nClaude 4.5 Haiku,0.16183333333333333,/models/claude-4-5-haiku-reasoning/providers,false\nLlama Nemotron Super 49B v1.5,0.16033333333333333,/models/llama-nemotron-super-49b-v1-5-reasoning/providers,false\nCommand A,0.15466666666666667,/models/command-a/providers,false\ngpt-oss-20B (high),0.14666666666666667,/models/gpt-oss-20b/providers,false\ngpt-oss-20B (low),0.13733333333333334,/models/gpt-oss-20b-low/providers,false\nGranite 4.0 H Small,0.1345,/models/granite-4-0-h-small/providers,false\nEXAONE 4.0 32B,0.133,/models/exaone-4-0-32b-reasoning/providers,false\nQwen3 8B,0.12733333333333333,/models/qwen3-8b-instruct-reasoning/providers,false\nNVIDIA Nemotron Nano 9B V2,0.10716666666666666,/models/nvidia-nemotron-nano-9b-v2-reasoning/providers,false\nQwen3 8B,0.10283333333333333,/models/qwen3-8b-instruct/providers,false"}
AA-Omniscience Hallucination Rate
AA-Omniscience Hallucination Rate (lower is better) measures how often the model answers incorrectly when it should have refused, defined as the proportion of wrong answers out of all non-correct attempts
{"@context":"https://schema.org","@type":"Dataset","name":"AA-Omniscience Hallucination Rate","creator":{"@type":"Organization","name":"Artificial Analysis","url":"https://artificialanalysis.ai"},"description":"AA-Omniscience Hallucination Rate (lower is better) measures how often the model answers incorrectly when it should have refused, defined as the proportion of wrong answers out of all non-correct attempts","measurementTechnique":"Independent test run by Artificial Analysis on dedicated hardware.","spatialCoverage":"Worldwide","keywords":["analytics","llm","AI","benchmark","model","gpt","claude"],"license":"https://creativecommons.org/licenses/by/4.0/","isAccessibleForFree":true,"citation":"Artificial Analysis (2025). LLM benchmarks dataset. https://artificialanalysis.ai","data":"modelName,omniscienceHallucinationRate,detailsUrl,isLabClaimedValue\nQwen3 8B,0.955043655953929,/models/qwen3-8b-instruct/providers,false\ngpt-oss-20B (high),0.932421875,/models/gpt-oss-20b/providers,false\nGLM-4.6,0.9308879445314248,/models/glm-4-6-reasoning/providers,false\nGPT-4.1,0.9228861330326945,/models/gpt-4-1/providers,false\ngpt-oss-120B (low),0.9051109753614335,/models/gpt-oss-120b-low/providers,false\nQwen3 8B,0.9035523300229182,/models/qwen3-8b-instruct-reasoning/providers,false\ngpt-oss-120B (high),0.8987710893563841,/models/gpt-oss-120b/providers,false\nQwen3 235B A22B 2507,0.8964262786218703,/models/qwen3-235b-a22b-instruct-2507-reasoning/providers,false\nGPT-5.1,0.891735918744229,/models/gpt-5-1-non-reasoning/providers,false\nMiniMax-M2,0.8888421052631579,/models/minimax-m2/providers,false\nGemini 2.5 Pro,0.8866968808317782,/models/gemini-2-5-pro/providers,false\nGemini 2.5 Flash (Sep),0.883131705090162,/models/gemini-2-5-flash-preview-09-2025-reasoning/providers,false\nGemini 3 Pro Preview,0.8798993167925206,/models/gemini-3-pro/providers,false\nGPT-5 (minimal),0.8777192580719029,/models/gpt-5-minimal/providers,false\nLlama 4 Maverick,0.8752720940356987,/models/llama-4-maverick/providers,false\nGranite 4.0 H Small,0.8725207009435779,/models/granite-4-0-h-small/providers,false\nQwen3 Next 80B A3B,0.8665987780040734,/models/qwen3-next-80b-a3b-reasoning/providers,false\nEXAONE 4.0 32B,0.8612072279892349,/models/exaone-4-0-32b-reasoning/providers,false\ngpt-oss-20B (low),0.8605100463678517,/models/gpt-oss-20b-low/providers,false\nDeepSeek R1 0528,0.8336082960169692,/models/deepseek-r1/providers,false\nGPT-5 (medium),0.8163428267234496,/models/gpt-5-medium/providers,false\nGPT-5 (high),0.8099375509095845,/models/gpt-5/providers,false\nDeepSeek V3.2 Exp,0.8060246462802373,/models/deepseek-v3-2-reasoning/providers,false\nCommand A,0.7791798107255521,/models/command-a/providers,false\nGPT-5 (low),0.7742864624247185,/models/gpt-5-low/providers,false\nQwen3 235B 2507,0.7626844552253891,/models/qwen3-235b-a22b-instruct-2507/providers,false\nLlama Nemotron Super 49B v1.5,0.7562524811433108,/models/llama-nemotron-super-49b-v1-5-reasoning/providers,false\nKimi K2 Thinking,0.7439943476212906,/models/kimi-k2-thinking/providers,false\nDeepSeek V3.1 Terminus,0.7395881006864988,/models/deepseek-v3-1-terminus-reasoning/providers,false\nGrok 4.1 Fast,0.716586852416195,/models/grok-4-1-fast-reasoning/providers,false\nKimi K2 0905,0.6895568231680561,/models/kimi-k2-0905/providers,false\nGrok 4 Fast,0.6714346712211785,/models/grok-4-fast-reasoning/providers,false\nGrok 4,0.6379405286343612,/models/grok-4/providers,false\nNVIDIA Nemotron Nano 9B V2,0.6027627403397424,/models/nvidia-nemotron-nano-9b-v2-reasoning/providers,false\nMagistral Medium 1.2,0.5970802919708029,/models/magistral-medium-2509/providers,false\nGPT-5 mini (high),0.5527909995672868,/models/gpt-5-mini/providers,false\nGPT-5.1 (high),0.5115919629057187,/models/gpt-5-1/providers,false\nLlama 3.1 405B,0.5110732538330494,/models/llama-3-1-instruct-405b/providers,false\nClaude 4.1 Opus,0.4838709677419355,/models/claude-4-1-opus-thinking/providers,false\nClaude 4.5 Sonnet,0.47732754462132176,/models/claude-4-5-sonnet-thinking/providers,false\nClaude 4.5 Haiku,0.2606880095446411,/models/claude-4-5-haiku-reasoning/providers,false"}
Intelligence Index Comparisons
Intelligence vs. Price
While higher intelligence models are typically more expensive, they do not all follow the same price-quality curve.
Combination metric covering multiple dimensions of intelligence - the simplest way to compare how smart models are. Version 3.0 was released in September 2025 and includes: MMLU-Pro, GPQA Diamond, Humanity's Last Exam, LiveCodeBench, SciCode, AIME 2025, IFBench, AA-LCR, Terminal-Bench Hard, 𝜏²-Bench Telecom. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.
Price per token, represented as USD per million Tokens. Price is a blend of Input & Output token prices (3:1 ratio).
Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).
Intelligence vs. Output Speed
There is a trade-off between model quality and output speed, with higher intelligence models typically having lower output speed.
Combination metric covering multiple dimensions of intelligence - the simplest way to compare how smart models are. Version 3.0 was released in September 2025 and includes: MMLU-Pro, GPQA Diamond, Humanity's Last Exam, LiveCodeBench, SciCode, AIME 2025, IFBench, AA-LCR, Terminal-Bench Hard, 𝜏²-Bench Telecom. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.
Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).
Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).
Intelligence vs. End-to-End Response Time
Combination metric covering multiple dimensions of intelligence - the simplest way to compare how smart models are. Version 3.0 was released in September 2025 and includes: MMLU-Pro, GPQA Diamond, Humanity's Last Exam, LiveCodeBench, SciCode, AIME 2025, IFBench, AA-LCR, Terminal-Bench Hard, 𝜏²-Bench Telecom. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.
Seconds to receive a 500 token response. Key components:
- Input time: Time to receive the first response token
- Thinking time (only for reasoning models): Time reasoning models spend outputting tokens to reason prior to providing an answer. Amount of tokens based on the average reasoning tokens across a diverse set of 60 prompts (methodology details).
- Answer time: Time to generate 500 output tokens, based on output speed
Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).
Intelligence Index Token Use & Cost
Output Tokens Used to Run Artificial Analysis Intelligence Index
The number of tokens required to run all evaluations in the Artificial Analysis Intelligence Index (excluding repeats).
Intelligence vs. Output Tokens Used in Artificial Analysis Intelligence Index
The number of tokens required to run all evaluations in the Artificial Analysis Intelligence Index (excluding repeats).
Combination metric covering multiple dimensions of intelligence - the simplest way to compare how smart models are. Version 3.0 was released in September 2025 and includes: MMLU-Pro, GPQA Diamond, Humanity's Last Exam, LiveCodeBench, SciCode, AIME 2025, IFBench, AA-LCR, Terminal-Bench Hard, 𝜏²-Bench Telecom. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.
Cost to Run Artificial Analysis Intelligence Index
The cost to run the evaluations in the Artificial Analysis Intelligence Index, calculated using the model's input and output token pricing and the number of tokens used across evaluations (excluding repeats).
Intelligence vs. Cost to Run Artificial Analysis Intelligence Index
The cost to run the evaluations in the Artificial Analysis Intelligence Index, calculated using the model's input and output token pricing and the number of tokens used across evaluations (excluding repeats).
Combination metric covering multiple dimensions of intelligence - the simplest way to compare how smart models are. Version 3.0 was released in September 2025 and includes: MMLU-Pro, GPQA Diamond, Humanity's Last Exam, LiveCodeBench, SciCode, AIME 2025, IFBench, AA-LCR, Terminal-Bench Hard, 𝜏²-Bench Telecom. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.
Context Window
Context Window
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).
{"@context":"https://schema.org","@type":"Dataset","name":"Context Window","creator":{"@type":"Organization","name":"Artificial Analysis","url":"https://artificialanalysis.ai"},"description":"Context Window: Tokens Limit; Higher is better","measurementTechnique":"Independent test run by Artificial Analysis on dedicated hardware.","spatialCoverage":"Worldwide","keywords":["analytics","llm","AI","benchmark","model","gpt","claude"],"license":"https://creativecommons.org/licenses/by/4.0/","isAccessibleForFree":true,"citation":"Artificial Analysis (2025). LLM benchmarks dataset. https://artificialanalysis.ai","data":""}
Intelligence vs. Context Window
Combination metric covering multiple dimensions of intelligence - the simplest way to compare how smart models are. Version 3.0 was released in September 2025 and includes: MMLU-Pro, GPQA Diamond, Humanity's Last Exam, LiveCodeBench, SciCode, AIME 2025, IFBench, AA-LCR, Terminal-Bench Hard, 𝜏²-Bench Telecom. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.
Maximum number of combined input & output tokens. Output tokens commonly have a significantly lower limit (varied by model).
Pricing
Pricing: Input and Output Prices
Price per token included in the request/message sent to the API, represented as USD per million Tokens.
Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).
Pricing: Cached Input Prompts
Price per token included in the request/message sent to the API, represented as USD per million Tokens.
One-time cost charged when storing a prompt in the cache for future reuse, represented as USD per million tokens.
Price per token for cached prompts (previously processed), typically offering a significant discount compared to regular input price, represented as USD per million tokens.
Cost to maintain tokens in cache storage, charged per million tokens per hour. Currently only applicable to Google's Gemini models.
Price per token generated by the model (received from the API), represented as USD per million Tokens.
Pricing: Image Input Pricing
Price for 1,000 images at a resolution of 1 Megapixel (1024 x 1024) processed by the model.
Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).
Intelligence vs. Price (Log Scale)
While higher intelligence models are typically more expensive, they do not all follow the same price-quality curve.
Combination metric covering multiple dimensions of intelligence - the simplest way to compare how smart models are. Version 3.0 was released in September 2025 and includes: MMLU-Pro, GPQA Diamond, Humanity's Last Exam, LiveCodeBench, SciCode, AIME 2025, IFBench, AA-LCR, Terminal-Bench Hard, 𝜏²-Bench Telecom. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.
Price per token, represented as USD per million Tokens. Price is a blend of Input & Output token prices (3:1 ratio).
Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).
Speed
Measured by Output Speed (tokens per second)
Output Speed
Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).
Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).
{"@context":"https://schema.org","@type":"Dataset","name":"Output Speed","creator":{"@type":"Organization","name":"Artificial Analysis","url":"https://artificialanalysis.ai"},"description":"Output Tokens per Second; Higher is better","measurementTechnique":"Independent test run by Artificial Analysis on dedicated hardware.","spatialCoverage":"Worldwide","keywords":["analytics","llm","AI","benchmark","model","gpt","claude"],"license":"https://creativecommons.org/licenses/by/4.0/","isAccessibleForFree":true,"citation":"Artificial Analysis (2025). LLM benchmarks dataset. https://artificialanalysis.ai","data":""}
Output Speed by Input Token Count (Context Length)
Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).
Length of tokens provided in the request. See Prompt Options above to see benchmarks of different input prompt lengths across other charts.
Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).
Output Speed Variance
Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).
Output Speed vs. Price
Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).
Price per token, represented as USD per million Tokens. Price is a blend of Input & Output token prices (3:1 ratio).
Latency vs. Output Speed
Tokens per second received while the model is generating tokens (ie. after first chunk has been received from the API for models which support streaming).
Time to first token received, in seconds, after API request sent. For reasoning models which share reasoning tokens, this will be the first reasoning token. For models which do not support streaming, this represents time to receive the completion.
Price per token, represented as USD per million Tokens. Price is a blend of Input & Output token prices (3:1 ratio).
Latency
Measured by Time (seconds) to First Token
Latency: Time To First Answer Token
Time to first answer token received, in seconds, after API request sent. For reasoning models, this includes the 'thinking' time of the model before providing an answer. For models which do not support streaming, this represents time to receive the completion.
Latency: Time To First Token
Time to first token received, in seconds, after API request sent. For reasoning models which share reasoning tokens, this will be the first reasoning token. For models which do not support streaming, this represents time to receive the completion.
Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).
{"@context":"https://schema.org","@type":"Dataset","name":"Latency: Time To First Token","creator":{"@type":"Organization","name":"Artificial Analysis","url":"https://artificialanalysis.ai"},"description":"Seconds to First Token Received; Lower is better","measurementTechnique":"Independent test run by Artificial Analysis on dedicated hardware.","spatialCoverage":"Worldwide","keywords":["analytics","llm","AI","benchmark","model","gpt","claude"],"license":"https://creativecommons.org/licenses/by/4.0/","isAccessibleForFree":true,"citation":"Artificial Analysis (2025). LLM benchmarks dataset. https://artificialanalysis.ai","data":""}
Time to First Token by Input Token Count (Context Length)
Length of tokens provided in the request. See Prompt Options above to see benchmarks of different input prompt lengths across other charts.
Time to first token received, in seconds, after API request sent. For reasoning models which share reasoning tokens, this will be the first reasoning token. For models which do not support streaming, this represents time to receive the completion.
Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).
Time to First Token Variance
Time to first token received, in seconds, after API request sent. For reasoning models which share reasoning tokens, this will be the first reasoning token. For models which do not support streaming, this represents time to receive the completion.
End-to-End Response Time
Seconds to output 500 Tokens, calculated based on time to first token, 'thinking' time for reasoning models, and output speed
End-to-End Response Time
Seconds to receive a 500 token response. Key components:
- Input time: Time to receive the first response token
- Thinking time (only for reasoning models): Time reasoning models spend outputting tokens to reason prior to providing an answer. Amount of tokens based on the average reasoning tokens across a diverse set of 60 prompts (methodology details).
- Answer time: Time to generate 500 output tokens, based on output speed
Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).
End-to-End Response Time by Input Token Count (Context Length)
Length of tokens provided in the request. See Prompt Options above to see benchmarks of different input prompt lengths across other charts.
Seconds to receive a 500 token response considering input processing time, 'thinking' time of reasoning models, and output speed.
Figures represent performance of the model's first-party API (e.g. OpenAI for o1) or the median across providers where a first-party API is not available (e.g. Meta's Llama models).
Model Size (Open Weights Models Only)
Model Size: Total and Active Parameters
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
Combination metric covering multiple dimensions of intelligence - the simplest way to compare how smart models are. Version 3.0 was released in September 2025 and includes: MMLU-Pro, GPQA Diamond, Humanity's Last Exam, LiveCodeBench, SciCode, AIME 2025, IFBench, AA-LCR, Terminal-Bench Hard, 𝜏²-Bench Telecom. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.
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. Total Parameters
Combination metric covering multiple dimensions of intelligence - the simplest way to compare how smart models are. Version 3.0 was released in September 2025 and includes: MMLU-Pro, GPQA Diamond, Humanity's Last Exam, LiveCodeBench, SciCode, AIME 2025, IFBench, AA-LCR, Terminal-Bench Hard, 𝜏²-Bench Telecom. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.
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.
Anthropic
Mistral
Ministral 8B
Perplexity
R1 1776
Amazon
Microsoft Azure
Liquid AI
Upstage
NVIDIA
Moonshot AI
Allen Institute for AI
Reka AI
Nous Research
DeepHermes 3 - Mistral 24B Preview (Non-reasoning)
LG AI Research
EXAONE 4.0 32B (Non-reasoning)
Baidu
Deep Cogito
Cogito v2.1 (Reasoning)
Z AI
Cohere
AI21 Labs
Alibaba
InclusionAI
Ring-flash-2.0
ByteDance Seed
OpenChat
OpenChat 3.5 (1210)
Databricks
DBRX Instruct
Snowflake
01.AI
Yi-LargeModels compared: OpenAI: GPT 4o Audio, GPT 4o Realtime, GPT 4o Speech Pipeline, GPT Realtime, GPT Realtime Mini (Oct '25), GPT-3.5 Turbo, GPT-3.5 Turbo (0125), GPT-3.5 Turbo (0301), GPT-3.5 Turbo (0613), GPT-3.5 Turbo (1106), GPT-3.5 Turbo Instruct, GPT-4, GPT-4 Turbo, GPT-4 Turbo (0125), GPT-4 Turbo (1106), GPT-4 Vision, GPT-4.1, GPT-4.1 mini, GPT-4.1 nano, GPT-4.5 (Preview), GPT-4o (Apr), GPT-4o (Aug), GPT-4o (ChatGPT), GPT-4o (Mar), GPT-4o (May), GPT-4o (Nov), GPT-4o Realtime (Dec), GPT-4o mini, GPT-4o mini Realtime (Dec), GPT-5 (ChatGPT), GPT-5 (high), GPT-5 (low), GPT-5 (medium), GPT-5 (minimal), GPT-5 Codex (high), GPT-5 Pro (high), GPT-5 mini (high), GPT-5 mini (medium), GPT-5 mini (minimal), GPT-5 nano (high), GPT-5 nano (medium), GPT-5 nano (minimal), GPT-5.1, GPT-5.1 (high), GPT-5.1 Codex (high), GPT-5.1 Codex mini (high), gpt-oss-120B (high), gpt-oss-120B (low), gpt-oss-20B (high), gpt-oss-20B (low), o1, o1-mini, o1-preview, o1-pro, o3, o3-mini, o3-mini (high), o3-pro, and o4-mini (high), Meta: Code Llama 70B, Llama 2 Chat 13B, Llama 2 Chat 70B, Llama 2 Chat 7B, Llama 3 70B, Llama 3 8B, Llama 3.1 405B, Llama 3.1 70B, Llama 3.1 8B, Llama 3.2 11B (Vision), Llama 3.2 1B, Llama 3.2 3B, Llama 3.2 90B (Vision), Llama 3.3 70B, Llama 4 Behemoth, Llama 4 Maverick, Llama 4 Scout, and Llama 65B, Google: Gemini 1.0 Pro, Gemini 1.0 Ultra, Gemini 1.5 Flash (May), Gemini 1.5 Flash (Sep), Gemini 1.5 Flash-8B, Gemini 1.5 Pro (May), Gemini 1.5 Pro (Sep), Gemini 2.0 Flash, Gemini 2.0 Flash (exp), Gemini 2.0 Flash Thinking exp. (Dec), Gemini 2.0 Flash Thinking exp. (Jan), Gemini 2.0 Flash-Lite (Feb), Gemini 2.0 Flash-Lite (Preview), Gemini 2.0 Pro Experimental, Gemini 2.5 Flash, Gemini 2.5 Flash Live Preview, Gemini 2.5 Flash Native Audio, Gemini 2.5 Flash Native Audio Dialog, Gemini 2.5 Flash (Sep), Gemini 2.5 Flash-Lite, Gemini 2.5 Flash-Lite (Sep), Gemini 2.5 Pro, Gemini 2.5 Pro (Mar), Gemini 2.5 Pro (May), Gemini 3 Pro Preview, Gemini 3 Pro Preview (low), Gemini Experimental (Nov), Gemma 2 27B, Gemma 2 2B, Gemma 2 9B, Gemma 3 12B, Gemma 3 1B, Gemma 3 270M, Gemma 3 27B, Gemma 3 4B, Gemma 3n E2B, Gemma 3n E4B, Gemma 3n E4B (May), Gemma 7B, PALM-2, and Whisperwind, Anthropic: Claude 2.0, Claude 2.1, Claude 3 Haiku, Claude 3 Opus, Claude 3 Sonnet, Claude 3.5 Haiku, Claude 3.5 Sonnet (June), Claude 3.5 Sonnet (Oct), Claude 3.7 Sonnet, Claude 4 Opus, Claude 4 Sonnet, Claude 4.1 Opus, Claude 4.5 Haiku, Claude 4.5 Sonnet, and Claude Instant, Mistral: Codestral (Jan), Codestral (May), Codestral-Mamba, Devstral Medium, Devstral Small, Devstral Small (May), Magistral Medium 1, Magistral Medium 1.1, Magistral Medium 1.2, Magistral Small 1, Magistral Small 1.1, Magistral Small 1.2, Ministral 3B, Ministral 8B, Mistral 7B, Mistral Large (Feb), Mistral Large 2 (Jul), Mistral Large 2 (Nov), Mistral Medium, Mistral Medium 3, Mistral Medium 3.1, Mistral NeMo, Mistral Saba, Mistral Small (Feb), Mistral Small (Sep), Mistral Small 3, Mistral Small 3.1, Mistral Small 3.2, Mixtral 8x22B, Mixtral 8x7B, Pixtral 12B, and Pixtral Large, DeepSeek: DeepSeek Coder V2 Lite, DeepSeek LLM 67B (V1), DeepSeek Prover V2 671B, DeepSeek R1 (FP4), DeepSeek R1 (Jan), DeepSeek R1 0528, DeepSeek R1 0528 Qwen3 8B, DeepSeek R1 Distill Llama 70B, DeepSeek R1 Distill Llama 8B, DeepSeek R1 Distill Qwen 1.5B, DeepSeek R1 Distill Qwen 14B, DeepSeek R1 Distill Qwen 32B, DeepSeek V3 (Dec), DeepSeek V3 0324, DeepSeek V3.1, DeepSeek V3.1 Terminus, DeepSeek V3.2 Exp, DeepSeek-Coder-V2, DeepSeek-OCR, DeepSeek-V2, DeepSeek-V2.5, DeepSeek-V2.5 (Dec), DeepSeek-VL2, and Janus Pro 7B, Perplexity: PPLX-70B Online, PPLX-7B-Online, R1 1776, Sonar, Sonar 3.1 Huge, Sonar 3.1 Large, Sonar 3.1 Small , Sonar Large, Sonar Pro, Sonar Reasoning, Sonar Reasoning Pro, and Sonar Small, xAI: Grok 2, Grok 3, Grok 3 Reasoning Beta, Grok 3 mini, Grok 3 mini Reasoning (low), Grok 3 mini Reasoning (high), Grok 4, Grok 4 Fast, Grok 4 Fast 1111 (Reasoning), Grok 4 mini (0908), Grok 4.1 Fast, Grok 4.1 Fast v4, Grok Beta, Grok Code Fast 1, Grok-1, and test model, OpenChat: OpenChat 3.5, Amazon: Nova 2.0 Lite, Nova 2.0 Lite (high), Nova 2.0 Lite (low), Nova 2.0 Lite (medium), Nova Lite, Nova Micro, Nova Premier, and Nova Pro, Microsoft Azure: Phi-3 Medium 14B, Phi-3 Mini, Phi-4, Phi-4 Mini, Phi-4 Multimodal, Phi-4 mini reasoning, Phi-4 reasoning, Phi-4 reasoning plus, and Yosemite, Liquid AI: LFM 1.3B, LFM 3B, LFM 40B, LFM2 1.2B, LFM2 2.6B, and LFM2 8B A1B, Upstage: Solar Mini, Solar Pro, Solar Pro (Nov), Solar Pro 2, and Solar Pro 2 , Databricks: DBRX, MiniMax: MiniMax M1 40k, MiniMax M1 80k, MiniMax-M2, and MiniMax-Text-01, NVIDIA: Cosmos Nemotron 34B, Llama 3.1 Nemotron 70B, Llama 3.1 Nemotron Nano 4B v1.1, Llama 3.1 Nemotron Nano 8B, Llama 3.3 Nemotron Nano 8B, Llama Nemotron Ultra, Llama 3.3 Nemotron Super 49B, Llama Nemotron Super 49B v1.5, NVIDIA Nemotron Nano 12B v2 VL, and NVIDIA Nemotron Nano 9B V2, StepFun: Step-2, Step-2-Mini, Step3, step-1-128k, step-1-256k, step-1-32k, step-1-8k, step-1-flash, step-2-16k-exp, and step-r1-v-mini, IBM: Granite 3.0 2B, Granite 3.0 8B, Granite 3.3 8B, Granite 4.0 1B, Granite 4.0 350M, Granite 4.0 8B, Granite 4.0 H 1B, Granite 4.0 H 350M, Granite 4.0 H Small, Granite 4.0 Micro, Granite 4.0 Tiny, and Granite Vision 3.3 2B, Inceptionlabs: Mercury, Mercury Coder Mini, Mercury Coder Small, and Mercury Instruct, Reka AI: Reka Core, Reka Edge, Reka Flash (Feb), Reka Flash, Reka Flash 3, and Reka Flash 3.1, LG AI Research: EXAONE 4.0 32B, EXAONE Deep 32B, and Exaone 4.0 1.2B, Xiaomi: MiMo 7B RL, Baidu: ERNIE 4.5, ERNIE 4.5 0.3B, ERNIE 4.5 21B A3B, ERNIE 4.5 300B A47B, ERNIE 4.5 VL 28B A3B, ERNIE 4.5 VL 424B A47B, ERNIE 5.0 Thinking Preview, and ERNIE X1, Baichuan: Baichuan 4 and Baichuan M1 (Preview), vercel: v0-1.0-md, Apple: Apple On-Device and FastVLM, Other: LLaVA-v1.5-7B, Tencent: Hunyuan A13B, Hunyuan 80B A13B, Hunyuan T1, and Hunyuan-TurboS, Z AI: GLM-4 32B, GLM-4 9B, GLM-4-Air, GLM-4 AirX, GLM-4 FlashX, GLM-4-Long, GLM-4-Plus, GLM-4.1V 9B Thinking, GLM-4.5, GLM-4.5-Air, GLM-4.5V, GLM-4.6, GLM-Z1 32B, GLM-Z1 9B, GLM-Z1 Rumination 32B, and GLM-Zero (Preview), Cohere: Aya Expanse 32B, Aya Expanse 8B, Command, Command A, Command Light, Command R7B, Command-R, Command-R (Mar), Command-R+ (Apr), and Command-R+, Bytedance: Duobao 1.5 Pro, Seed-Thinking-v1.5, Skylark Lite, and Skylark Pro, AI21 Labs: Jamba 1.5 Large, Jamba 1.5 Large (Feb), Jamba 1.5 Mini, Jamba 1.5 Mini (Feb), Jamba 1.6 Large, Jamba 1.6 Mini, Jamba 1.7 Large, Jamba 1.7 Mini, Jamba Instruct, and Jamba Reasoning 3B, Snowflake: Arctic and Snowflake Llama 3.3 70B, PaddlePaddle: PaddleOCR-VL-0.9B, Alibaba: QwQ-32B, QwQ 32B-Preview, Qwen Chat 14B, Qwen Chat 72B, Qwen Chat 7B, Qwen1.5 Chat 110B, Qwen1.5 Chat 14B, Qwen1.5 Chat 32B, Qwen1.5 Chat 72B, Qwen1.5 Chat 7B, Qwen2 72B, Qwen2 Instruct 7B, Qwen2 Instruct A14B 57B, Qwen2-VL 72B, Qwen2.5 Coder 32B, Qwen2.5 Coder 7B , Qwen2.5 Instruct 14B, Qwen2.5 Instruct 32B, Qwen2.5 72B, Qwen2.5 Instruct 7B, Qwen2.5 Max, Qwen2.5 Max 01-29, Qwen2.5 Omni 7B, Qwen2.5 Plus, Qwen2.5 Turbo, Qwen2.5 VL 72B, Qwen2.5 VL 7B, Qwen3 0.6B, Qwen3 1.7B, Qwen3 14B, Qwen3 235B, Qwen3 235B A22B 2507, Qwen3 235B 2507, Qwen3 30B, Qwen3 30B A3B 2507, Qwen3 32B, Qwen3 4B, Qwen3 4B 2507, Qwen3 8B, Qwen3 Coder 30B A3B, Qwen3 Coder 480B, Qwen3 Max, Qwen3 Max (Preview), Qwen3 Max Thinking, Qwen3 Next 80B A3B, Qwen3 Omni 30B A3B, Qwen3 VL 235B A22B, Qwen3 VL 30B A3B, Qwen3 VL 32B, Qwen3 VL 4B, and Qwen3 VL 8B, InclusionAI: Ling-1T, Ling-flash-2.0, Ling-mini-2.0, Ring-1T, and Ring-flash-2.0, 01.AI: Yi-Large and Yi-Lightning, and ByteDance Seed: Doubao Seed Code and Seed-OSS-36B-Instruct.