Comparación de modelos: análisis de inteligencia, rendimiento y precio

Entorno de pruebas Microevals
Comparación y análisis de modelos de IA según métricas clave de rendimiento, como calidad, precio, velocidad de salida, latencia, ventana de contexto y otras. Haz clic en cualquier modelo para ver métricas detalladas. Para obtener más información, incluida nuestra metodología, consulta nuestras preguntas frecuentes.

Inteligencia

Logotipo de Claude Opus 5 (max) Claude Opus 5 (max) y Logotipo de Claude Opus 5 (xhigh) Claude Opus 5 (xhigh) son los modelos con mayor inteligencia, seguidos de Logotipo de Claude Fable 5 (with fallback) Claude Fable 5 (with fallback) y Logotipo de Claude Opus 5 (high) Claude Opus 5 (high).

Velocidad de salida (tokens/s)

Logotipo de Celeris-1 Celeris-1 (1513 t/s) y Logotipo de Mercury 2 Mercury 2 (940 t/s) son los modelos más rápidos, seguidos de Logotipo de Gemini 3.5 Flash-Lite Gemini 3.5 Flash-Lite y Logotipo de LFM2.5-VL-1.6B LFM2.5-VL-1.6B.

Latencia (segundos)

Logotipo de Gemini 2.5 Flash-Lite Gemini 2.5 Flash-Lite (0.29s) y Logotipo de Command A+ Command A+ (0.41s) son los modelos con menor latencia, seguidos de Logotipo de Gemini 2.5 Flash Gemini 2.5 Flash y Logotipo de North Mini Code North Mini Code.

Precio ($ por millón de tokens)

Logotipo de Devstral 2 Devstral 2 ($0.00) y Logotipo de North Mini Code North Mini Code ($0.00) son los modelos más económicos, seguidos de Logotipo de Gemma 3 4B Gemma 3 4B y Logotipo de Gemma 3 27B Gemma 3 27B.

Ventana de contexto

Logotipo de Llama 4 Scout Llama 4 Scout (10M) y Logotipo de Grok 4.20 0309 Grok 4.20 0309 (2M) son los modelos con las ventanas de contexto más grandes, seguidos de Logotipo de Gemini 1.5 Pro (May) Gemini 1.5 Pro (May) y Logotipo de Grok 4.1 Fast Grok 4.1 Fast.

Aspectos destacados

Inteligencia

Artificial Analysis Intelligence Index · Higher is better

Velocidad

Output tokens per second · Higher is better

Costo por tarea

Weighted average cost (USD) per Intelligence Index task · Lower is better

Inteligencia

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v4.1.1 incorporates 9 evaluations: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR
Reasoning models are indicated by a lightbulb icon

Artificial Analysis Intelligence Index v4.1.1 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.

Artificial Analysis Intelligence Index by Open Weights / Proprietary

Artificial Analysis Intelligence Index v4.1.1 incorporates 9 evaluations: GDPval-AA v2, 𝜏³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, AA-LCR
Reasoning models are indicated by a lightbulb icon

Artificial Analysis Intelligence Index v4.1.1 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.

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).

Benchmarks

Intelligence Evaluations

Intelligence evaluations measured independently by Artificial Analysis · Higher is better
See more
GDPval-AA v2

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

𝜏³-BankingUpdated

Agentic tool use

Terminal-Bench v2.1

Agentic coding & terminal use

SciCode

Coding

Humanity's Last ExamUpdated

Reasoning & knowledge

GPQA Diamond

Scientific reasoning

CritPt

Physics reasoning

AA-Omniscience AccuracyUpdated

Knowledge

AA-Omniscience Non-Hallucination RateUpdated

1 - hallucination rate

AA-LCRUpdated

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-AA

Agentic business operations

AA-AnalystAgentNew

Quantitative analysis on spreadsheets & documents

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.

Artificial Analysis Intelligence Index v4.1.1 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.

AA-Briefcase

AA-Briefcase Elo

AA-Briefcase is an agentic knowledge work benchmark developed by Artificial Analysis. AA-Briefcase Elo is a combined metric that aggregates rubric pass rate, analytical quality Elo and presentation Elo · Higher is better
Reasoning models are indicated by a lightbulb icon

AA-Briefcase Elo is a combined metric that aggregates analytical quality Elo, presentation Elo, and rubric pass rate, with rubric performance converted into Elo via synthetic head-to-head matches. Elo and 95% confidence interval bounds are clamped at 0.

AA-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.
Reasoning models are indicated by a lightbulb icon

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.

Índice de apertura

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

Comparaciones del Índice de Inteligencia

Intelligence Index vs. Cost per Intelligence Index Task

Artificial Analysis Intelligence Index · Weighted average cost (USD) per Artificial Analysis Intelligence Index task
Most attractive quadrant
Pareto line
Reasoning models are indicated by a lightbulb icon

Weighted average cost per Intelligence Index task. Each evaluation’s cost is calculated from input, cache hit, cache write, reasoning, and answer token prices, divided by task count, and weighted by its Intelligence Index weight.

Artificial Analysis Intelligence Index v4.1.1 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.

Uso de tokens

Output Tokens per Intelligence Index Task

Weighted average number of output tokens used to run one task in the Artificial Analysis Intelligence Index
Reasoning models are indicated by a lightbulb icon

The number of tokens required per Intelligence Index task. This is calculated by multiplying the output tokens per eval by the relative weights of each benchmark in the Intelligence Index, then dividing by task count (excluding repeats).

Costo

Cost per Intelligence Index Task

Weighted average cost (USD) per Artificial Analysis Intelligence Index task, segmented by token type. Lower is better
Reasoning models are indicated by a lightbulb icon

Weighted average cost per Intelligence Index task. Each evaluation’s cost is calculated from input, cache hit, cache write, reasoning, and answer token prices, divided by task count, and weighted by its Intelligence Index weight.

Cost to Run Artificial Analysis Intelligence Index

Cost (USD) to run all evaluations in the Artificial Analysis Intelligence Index
Reasoning models are indicated by a lightbulb icon

The cost to run the evaluations in the Artificial Analysis Intelligence Index, calculated using the model's input, cache hit, cache write, reasoning, and answer token prices and the number of tokens used across evaluations (excluding repeats).

Pricing: Cache Hit, Input, and Output

Price (USD per M Tokens)
Reasoning models are indicated by a lightbulb icon

Price per token for cached prompts (previously processed), typically offering a significant discount compared to regular input price, represented as USD per million tokens. The values shown here are the cache hit price; cache write and cache storage are billed separately and vary by provider — see "Cache pricing by provider" for detail.

Ventana de contexto

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).

Velocidad

Medida por la velocidad de salida (tokens por segundo)

Output Speed

Output tokens per second · Higher is better
Reasoning models are indicated by a lightbulb icon

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).

Time per Intelligence Index Task

Weighted average decode time (minutes) per task; excludes TTFT and overhead time · Lower is better
Reasoning models are indicated by a lightbulb icon

The weighted average time (seconds) per Artificial Analysis Intelligence Index task. This is calculated by dividing output tokens per task by output speed, weighted by the relative weights of each benchmark in the Intelligence Index.

Latencia

Medida por el tiempo (segundos) hasta el primer token

Latency: Time To First Answer Token

Seconds to first answer token received · Accounts for reasoning model 'thinking' time
Reasoning models are indicated by a lightbulb icon

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.

Tiempo de respuesta de extremo a extremo

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 output 500 tokens, including reasoning model 'thinking' time · Lower is better
Reasoning models are indicated by a lightbulb icon

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).

Tamaño del modelo (solo modelos de pesos abiertos)

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.

Preguntas frecuentes

Claude Opus 5 (Adaptive Reasoning, Max Effort) lidera actualmente el Índice de Inteligencia de Artificial Analysis con una puntuación de 63, entre 159 modelos evaluados.

Los mejores modelos de IA según el Índice de Inteligencia son: 1. Claude Opus 5 (Adaptive Reasoning, Max Effort) (63), 2. Claude Opus 5 (Adaptive Reasoning, Xhigh Effort) (63), 3. Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) (62), 4. Claude Opus 5 (Adaptive Reasoning, High Effort) (61) y 5. GPT-5.6 Sol (max) (61).

Celeris-1 es el más rápido, con 1,512.9 tokens por segundo, seguido de Mercury 2 (940.0 t/s) y Gemini 3.5 Flash-Lite (372.6 t/s).

Llama 3.1 Instruct 8B es el más económico, con $0.02 por millón de tokens (precio combinado), seguido de Nova Micro ($0.03) y Sarvam 30B (high) ($0.03).

Gemini 2.5 Flash-Lite (Non-reasoning) tiene el menor tiempo hasta el primer token, con 0.29s, seguido de Command A+ (0.41s) y Gemini 2.5 Flash (Non-reasoning) (0.48s).

Kimi K3 (max) es el modelo con pesos abiertos mejor clasificado, con una puntuación de 60 en el Índice de Inteligencia. Hay 84 modelos con pesos abiertos de un total de 159 evaluados.

Los mejores modelos de IA con pesos abiertos según el Índice de Inteligencia son: 1. Kimi K3 (max) (60), 2. Qwen3.8 2.4T A95B (58) y 3. DeepSeek V4 Pro 0813 (Reasoning, Max Effort) (53).

Claude Opus 5 (Adaptive Reasoning, Max Effort) lidera entre 126 modelos de razonamiento con una puntuación de 63 en el Índice de Inteligencia. Los modelos de razonamiento usan pensamiento extendido para resolver problemas complejos antes de responder.

Los modelos se comparan en múltiples dimensiones, como inteligencia (calidad), precio, velocidad de salida (tokens por segundo), latencia (tiempo hasta el primer token), tiempo de respuesta de extremo a extremo y tamaño de la ventana de contexto. Las métricas de rendimiento se miden directamente mediante prompts estandarizados en 608 modelos.

Haz clic en el nombre o la fila de cualquier modelo en los gráficos para ver su página específica, con métricas detalladas y comparaciones directas con modelos similares. También puedes usar el selector de modelos para personalizar cuáles aparecen en cada gráfico. Ver el ranking