Microsoft Azure: inteligencia, rendimiento y precio de sus modelos

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

Microsoft Azure

Análisis de los modelos de Microsoft Azure en métricas clave como calidad, precio, velocidad de salida, latencia, ventana de contexto y más. Este análisis está pensado para ayudarte a elegir el mejor modelo ofrecido por Microsoft Azure para tu caso de uso.

Más inteligente

Claude Fable 5 (with fallback)

Claude Opus 4.7 (max)

Claude Sonnet 5 (max)

GPT-5.4 (xhigh)

Claude Sonnet 4.6 (max)

Índice de inteligencia

68 modelos en total

Más rápido

gpt-oss-120b (low)

370 t/s

gpt-oss-120b (high)

322 t/s

GPT-4.1 nano

244 t/s

o3-mini (high)

218 t/s

o3-mini

216 t/s

Velocidad de salida

68 modelos en total

Menor precio

GPT-5 nano (high)

$0.05

GPT-5 nano (medium)

$0.05

GPT-4.1 nano

$0.08

GPT-4o mini

$0.14

gpt-oss-120b (high)

$0.20

Precio combinado (por 1M de tokens)

68 modelos en total

Indica un modelo de razonamiento

Azure ofrece 68 modelos, cada uno con distintas características de inteligencia, rendimiento y precio. A continuación se comparan las métricas clave entre modelos.

En inteligencia, los mejores modelos en Azure son Claude Fable 5 (with fallback) (60), Claude Opus 4.7 (max) (54) y Claude Sonnet 5 (max) (53).
En velocidad de salida, los modelos más rápidos son gpt-oss-120b (low) (370 t/s), gpt-oss-120b (high) (322 t/s) y GPT-4.1 nano (244 t/s). La velocidad varía significativamente entre modelos, con una diferencia del 71% entre el más rápido y el más lento.
En latencia, Phi-4 Multimodal (0.81s), Phi-4 Mini (0.83s) y GPT-5.4 mini (0.93s) ofrecen el menor tiempo hasta el primer token de respuesta.
En precios, GPT-5 nano (high) ($0.05), GPT-5 nano (medium) ($0.05) y GPT-4.1 nano ($0.08) ofrecen los menores precios combinados por 1M de tokens. Los precios varían hasta 3.6x entre modelos.
En tamaño de ventana de contexto, GPT-5.4 (xhigh) (1M), Claude Fable 5 (with fallback) (1M) y Claude Opus 4.7 (max) (1M) admiten las ventanas de contexto más grandes en Azure.

Aspectos destacados

Inteligencia

Artificial Analysis Intelligence Index · Higher is better

Velocidad

Output tokens per second · Higher is better

Precio

USD per 1M tokens (blended) · Lower is better

Evaluaciones de inteligencia

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.

Intelligence Index vs. Price

Blended at 7:2:1 (cache-input-output) · USD per 1M tokens (blended)

Most attractive quadrant

Reasoning models are indicated by a lightbulb icon

While higher intelligence models are typically more expensive, they do not all follow the same price-quality curve.

Price per token, shown in USD per million tokens. Price is a blend of cache hit, input, and output token prices using the selected ratio (default 7:2:1 cache-input-output).

The blended cache price shown here uses cache hit price only. Other caching costs differ by provider:

Anthropic: charges a separate cache write fee, with different rates for 5-minute and 1-hour TTLs (1-hour TTL is more expensive).
Google (Vertex/Gemini): charges a per-hour cache storage fee in addition to cache hit pricing. Some providers also use tiered pricing for prompts above 200K tokens.
OpenAI, DeepSeek, others: typically charge only cache hit pricing with no write or storage fee.

See Prompt Caching for the full breakdown.

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).

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).

Precios

Intelligence Index vs. Price

Blended at 7:2:1 (cache-input-output) · USD per 1M tokens (blended)

Most attractive quadrant

Reasoning models are indicated by a lightbulb icon

While higher intelligence models are typically more expensive, they do not all follow the same price-quality curve.

Price per token, shown in USD per million tokens. Price is a blend of cache hit, input, and output token prices using the selected ratio (default 7:2:1 cache-input-output).

The blended cache price shown here uses cache hit price only. Other caching costs differ by provider:

Anthropic: charges a separate cache write fee, with different rates for 5-minute and 1-hour TTLs (1-hour TTL is more expensive).
Google (Vertex/Gemini): charges a per-hour cache storage fee in addition to cache hit pricing. Some providers also use tiered pricing for prompts above 200K tokens.
OpenAI, DeepSeek, others: typically charge only cache hit pricing with no write or storage fee.

See Prompt Caching for the full breakdown.

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).

Resumen de rendimiento

Output Speed vs. Price

Output speed: output tokens per second · USD per 1M tokens (blended)

Most attractive quadrant

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).

Price per token, shown in USD per million tokens. Price is a blend of cache hit, input, and output token prices using the selected ratio (default 7:2:1 cache-input-output).

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).

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 vs. Price

End-to-end response time: end-to-end seconds to output 500 tokens · USD per 1M tokens (blended)

Most attractive quadrant

Reasoning models are indicated by a lightbulb icon

Price per token, shown in USD per million tokens. Price is a blend of cache hit, input, and output token prices using the selected ratio (default 7:2:1 cache-input-output).

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 median (P50) measurement over the past 72 hours to reflect sustained changes in performance.


									Análisis adicional
Claude Fable 5 (with fallback)	1M	Propietario	60	$8.41	67	100.32	107.75	--	Modelo Proveedor
Claude Opus 4.7 (max)	1M	Propietario	54	$1.79	51	11.04	20.83	--	Modelo Proveedor
Claude Sonnet 5 (max)	1M	Propietario	53	$1.46	77	231.96	238.43	--	Modelo Proveedor
GPT-5.4 (xhigh)	1.05M	Propietario	51	$0.78	120	174.24	178.40	--	Modelo Proveedor
Claude Sonnet 4.6 (max)	200k	Propietario	47	$1.18	54	110.45	119.79	--	Modelo Proveedor
DeepSeek V4 Pro (max)	1M	Abierto	44	$0.76	76	1.51	65.28	57.23	Modelo Proveedor
Kimi K2.6	262k	Abierto	44	$0.66	149	1.49	34.63	29.79	Modelo Proveedor
Claude Opus 4.6 (max)	1M	Propietario	44*	--	43	13.67	25.34	--	Modelo Proveedor
DeepSeek V4 Pro (high)	1M	Abierto	43	$0.85	80	1.63	32.91	25.01	Modelo Proveedor
GPT-5.2 (xhigh)	400k	Propietario	42*	--	85	132.36	138.22	--	Modelo Proveedor
Claude Opus 4.5	200k	Propietario	41*	--	52	13.35	22.95	--	Modelo Proveedor
GPT-5.2 Codex (xhigh)	400k	Propietario	40*	--	109	117.73	122.30	--	Modelo Proveedor
GPT-5.4 mini (xhigh)	400k	Propietario	40	$0.49	194	164.05	166.63	--	Modelo Proveedor
GPT-5.2 (medium)	400k	Propietario	38*	--	71	5.81	12.85	--	Modelo Proveedor
Claude Opus 4.6 (high)	1M	Propietario	38*	--	41	3.07	15.33	--	Modelo Proveedor
Grok 4.3 (high)	200k	Propietario	38	$0.46	102	26.57	31.48	--	Modelo Proveedor
Grok 4.20 0309 v2	262k	Propietario	37*	--	193	12.10	14.69	--	Modelo Proveedor
GPT-5.1 (high)	272k	Propietario	37	$0.27	132	33.01	36.79	--	Modelo Proveedor
Claude 4.5 Sonnet	200k	Propietario	36	$1.06	45	11.21	22.22	--	Modelo Proveedor
GPT-5 Codex (high)	400k	Propietario	36*	--	140	11.90	15.48	--	Modelo Proveedor
Grok 4.3 (medium)	200k	Propietario	36*	--	95	15.17	20.42	--	Modelo Proveedor
Claude Sonnet 4.6 (Non-reasoning)	200k	Propietario	36*	--	44	2.08	13.33	--	Modelo Proveedor
Grok 4.3 (low)	200k	Propietario	35*	--	94	7.55	12.88	--	Modelo Proveedor
Kimi K2.5	262k	Abierto	35	$0.23	143	1.35	25.62	20.77	Modelo Proveedor
GPT-5.1 Codex (high)	400k	Propietario	35*	--	109	9.07	13.65	--	Modelo Proveedor
GPT-5 (high)	400k	Propietario	35	$0.21	80	80.21	86.47	--	Modelo Proveedor
Claude Opus 4.5	200k	Propietario	35*	--	46	1.89	12.68	--	Modelo Proveedor
Kimi K2.6	262k	Abierto	35*	--	177	1.75	4.57	--	Modelo Proveedor
GPT-5 (medium)	400k	Propietario	34*	--	82	31.20	37.30	--	Modelo Proveedor
Claude 4.1 Opus	200k	Propietario	34*	--	--	--	--	--	Modelo Proveedor
Grok 4	256k	Propietario	33*	--	57	10.24	19.09	--	Modelo Proveedor
Kimi K2 Thinking	256k	Abierto	33*	--	131	1.40	20.53	15.31	Modelo Proveedor
o3-pro	200k	Propietario	33*	--	--	--	--	--	Modelo Proveedor
DeepSeek V4 Pro	1M	Abierto	31*	--	83	1.76	7.75	--	Modelo Proveedor
GPT-5 (low)	400k	Propietario	31*	--	81	5.82	12.03	--	Modelo Proveedor
GPT-5 mini (medium)	400k	Propietario	31*	--	101	13.95	18.88	--	Modelo Proveedor
GPT-5.1 Codex mini (high)	400k	Propietario	31*	--	146	28.06	31.49	--	Modelo Proveedor
o3	200k	Propietario	30*	--	111	30.65	35.17	--	Modelo Proveedor
GPT-5.4 mini (medium)	400k	Propietario	30*	--	170	10.14	13.09	--	Modelo Proveedor
Claude 4.5 Haiku	200k	Propietario	30	$0.51	100	15.14	20.11	--	Modelo Proveedor
Kimi K2.5	262k	Abierto	29*	--	113	1.38	5.81	--	Modelo Proveedor
Claude 4.5 Sonnet	200k	Propietario	29*	--	41	2.18	14.34	--	Modelo Proveedor
Claude 4.1 Opus	200k	Propietario	28*	--	--	--	--	--	Modelo Proveedor
Grok 4 Fast	2M	Propietario	27*	--	--	--	--	--	Modelo Proveedor
GPT-5.2	400k	Propietario	26*	--	68	1.26	8.61	--	Modelo Proveedor
o4-mini (high)	200k	Propietario	26*	--	121	26.37	30.50	--	Modelo Proveedor
GPT-5 mini (high)	400k	Propietario	25	$0.03	101	96.24	101.19	--	Modelo Proveedor
DeepSeek V3.2	128k	Abierto	25*	--	145	1.89	5.34	--	Modelo Proveedor
gpt-oss-120b (high)	131k	Abierto	24	$0.06	322	0.82	8.59	6.21	Modelo Proveedor
Claude 4.5 Haiku	200k	Propietario	24*	--	91	1.22	6.72	--	Modelo Proveedor
o1	200k	Propietario	23*	--	--	--	--	--	Modelo Proveedor
Grok 3 mini Reasoning (high)	32k	Propietario	23*	--	--	--	--	--	Modelo Proveedor
GPT-5.1	400k	Propietario	20*	--	116	1.78	6.08	--	Modelo Proveedor
DeepSeek R1 0528	128k	Abierto	20*	--	--	--	--	--	Modelo Proveedor
GPT-5 nano (high)	400k	Propietario	20*	--	145	123.60	127.04	--	Modelo Proveedor
GPT-4.1	1M	Propietario	19*	--	88	1.40	7.10	--	Modelo Proveedor
GPT-5 nano (medium)	400k	Propietario	19*	--	125	60.35	64.36	--	Modelo Proveedor
o3-mini	200k	Propietario	19*	--	216	7.51	9.82	--	Modelo Proveedor
DeepSeek R1 (Jan)	128k	Abierto	19	$0.22	--	--	--	--	Modelo Proveedor
Grok 3	16k	Propietario	18*	--	--	--	--	--	Modelo Proveedor
GPT-5 (minimal)	400k	Propietario	17*	--	81	1.65	7.79	--	Modelo Proveedor
o1-preview	128k	Propietario	17*	--	--	--	--	--	Modelo Proveedor
GPT-5.4 mini	400k	Propietario	17*	--	150	0.93	4.26	--	Modelo Proveedor
Grok 4 Fast	2M	Propietario	16*	--	--	--	--	--	Modelo Proveedor
Mistral Large 3	256k	Abierto	16	$0.06	57	1.57	10.36	--	Modelo Proveedor
o3-mini (high)	200k	Propietario	16	--	218	25.51	27.80	--	Modelo Proveedor
gpt-oss-120b (low)	131k	Abierto	15	$0.02	370	1.07	7.83	5.41	Modelo Proveedor
GPT-4.1 mini	1M	Propietario	15	$0.05	94	1.46	6.77	--	Modelo Proveedor
Llama 4 Maverick (FP8)	128k	Abierto	14	$0.03	116	1.10	5.42	--	Modelo Proveedor
GPT-5 mini (minimal) East US 2 - Global Standard	400k	Propietario	14*	--	--	--	--	--	Modelo Proveedor
Mistral Medium 3	128k	Propietario	12*	--	--	--	--	--	Modelo Proveedor
GPT-4o (Nov)	128k	Propietario	11*	--	130	2.33	6.17	--	Modelo Proveedor
Llama 4 Scout	128k	Abierto	10	$0.01	--	--	--	--	Modelo Proveedor
GPT-4o (Aug)	128k	Propietario	10*	--	141	1.36	4.89	--	Modelo Proveedor
GPT-4.1 nano	1M	Propietario	10	$0.03	244	1.43	3.47	--	Modelo Proveedor
Llama 3.3 70B	128k	Abierto	9	$0.10	93	2.30	7.65	--	Modelo Proveedor
GPT-4o (May)	128k	Propietario	9*	--	141	1.41	4.95	--	Modelo Proveedor
GPT-5 nano (minimal) East US 2 - Global Standard	400k	Propietario	8*	--	--	--	--	--	Modelo Proveedor
GPT-4 Turbo	128k	Propietario	8*	--	100	1.67	6.69	--	Modelo Proveedor
Command A	256k	Abierto	8*	--	41	3.11	15.17	--	Modelo Proveedor
GPT-4o mini	128k	Propietario	7*	--	84	1.63	7.57	--	Modelo Proveedor
Phi-4 Mini	128k	Abierto	6	$0.00	44	0.83	12.15	--	Modelo Proveedor
Phi-4	16.4k	Abierto	5*	--	--	--	--	--	Modelo Proveedor
Phi-4 Multimodal	128k	Abierto	5*	--	18	0.81	29.21	--	Modelo Proveedor

Definiciones clave

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

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.

Average cost per task in the index. Costs are split by input, cache hit, cache write, reasoning, and answer token pricing where canonical token counts are available.

Price per token included in the request/message sent to the API, 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. 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.

Price per token to write prompt tokens into the cache so that later requests can hit them, represented as USD per million tokens. Some providers charge a premium over the standard input price to create a cache entry (e.g. Anthropic), while others cache automatically with no separate write fee.

Price per token generated by the model (received from the API), represented as USD per million Tokens.

Metrics are 'live' and are based on the past 72 hours of measurements, measurements are taken 8 times a day for single requests and 2 times per day for parallel requests.

Preguntas frecuentes

Preguntas comunes sobre Microsoft Azure

Microsoft Azure ofrece 68 modelos que monitoreamos: Claude Fable 5 (with fallback), Claude Opus 4.7 (max), Claude Sonnet 5 (max), GPT-5.4 (xhigh), Claude Sonnet 4.6 (max), DeepSeek V4 Pro (max), Kimi K2.6, Claude Opus 4.6 (max), DeepSeek V4 Pro (high), GPT-5.2 (xhigh), Claude Opus 4.5, GPT-5.2 Codex (xhigh), GPT-5.4 mini (xhigh), GPT-5.2 (medium), Claude Opus 4.6 (high), Grok 4.3 (high), Grok 4.20 0309 v2, GPT-5.1 (high), Claude 4.5 Sonnet, GPT-5 Codex (high), Grok 4.3 (medium), Claude Sonnet 4.6 (Non-reasoning), Grok 4.3 (low), Kimi K2.5, GPT-5.1 Codex (high), GPT-5 (high), Claude Opus 4.5, Kimi K2.6, GPT-5 (medium), Grok 4, Kimi K2 Thinking, DeepSeek V4 Pro, GPT-5 (low), GPT-5 mini (medium), GPT-5.1 Codex mini (high), o3, GPT-5.4 mini (medium), Claude 4.5 Haiku, Kimi K2.5, Claude 4.5 Sonnet, GPT-5.2, o4-mini (high), GPT-5 mini (high), DeepSeek V3.2, gpt-oss-120b (high), Claude 4.5 Haiku, GPT-5.1, GPT-5 nano (high), GPT-4.1, GPT-5 nano (medium), o3-mini, GPT-5 (minimal), GPT-5.4 mini, Mistral Large 3, o3-mini (high), gpt-oss-120b (low), GPT-4.1 mini, Llama 4 Maverick (FP8), GPT-4o (Nov), GPT-4o (Aug), GPT-4.1 nano, Llama 3.3 70B, GPT-4o (May), GPT-4 Turbo, Command A, GPT-4o mini, Phi-4 Mini y Phi-4 Multimodal.

El modelo más inteligente disponible en Microsoft Azure es Claude Fable 5 (with fallback), con una puntuación de 60 en el Índice de Inteligencia.

El modelo más rápido en Microsoft Azure por velocidad de salida es gpt-oss-120b (low), con 370.0 tokens por segundo.

El modelo con el menor tiempo hasta el primer token de respuesta en Microsoft Azure es Phi-4 Multimodal, con 0.81s. Una menor latencia significa una respuesta inicial más rápida.

El modelo más económico en Microsoft Azure por precio combinado es GPT-5 nano (high), a $0.05 por 1M de tokens (proporción 7:2:1 de aciertos de caché/entrada/salida).

Los precios en Microsoft Azure varían hasta 262x entre modelos, desde $0.05 por 1M de tokens para GPT-5 nano (high) hasta $14.00 por 1M de tokens para Claude Fable 5 (with fallback).

Sí, Microsoft Azure ofrece una API compatible con OpenAI, lo que facilita cambiar desde OpenAI o usar integraciones existentes del SDK de OpenAI.

55 de 68 modelos en Microsoft Azure admiten el modo JSON para salida estructurada.

67 de 68 modelos en Microsoft Azure admiten llamadas a funciones (uso de herramientas).

Sí, Microsoft Azure ofrece 41 modelos de razonamiento: Claude Fable 5 (with fallback), Claude Opus 4.7 (max), Claude Sonnet 5 (max), GPT-5.4 (xhigh), Claude Sonnet 4.6 (max), DeepSeek V4 Pro (max), Kimi K2.6, Claude Opus 4.6 (max), DeepSeek V4 Pro (high), GPT-5.2 (xhigh), Claude Opus 4.5, GPT-5.2 Codex (xhigh), GPT-5.4 mini (xhigh), GPT-5.2 (medium), Grok 4.3 (high), Grok 4.20 0309 v2, GPT-5.1 (high), Claude 4.5 Sonnet, GPT-5 Codex (high), Grok 4.3 (medium), Grok 4.3 (low), Kimi K2.5, GPT-5.1 Codex (high), GPT-5 (high), GPT-5 (medium), Grok 4, Kimi K2 Thinking, GPT-5 (low), GPT-5 mini (medium), GPT-5.1 Codex mini (high), o3, GPT-5.4 mini (medium), Claude 4.5 Haiku, o4-mini (high), GPT-5 mini (high), gpt-oss-120b (high), GPT-5 nano (high), GPT-5 nano (medium), o3-mini, o3-mini (high) y gpt-oss-120b (low). Los modelos de razonamiento usan pensamiento extendido para resolver problemas complejos antes de responder.

Sí, 17 de 68 modelos en Microsoft Azure son modelos de pesos abiertos: DeepSeek V4 Pro (max), Kimi K2.6, DeepSeek V4 Pro (high), Kimi K2.5, Kimi K2.6, Kimi K2 Thinking, DeepSeek V4 Pro, Kimi K2.5, DeepSeek V3.2, gpt-oss-120b (high), Mistral Large 3, gpt-oss-120b (low), Llama 4 Maverick (FP8), Llama 3.3 70B, Command A, Phi-4 Mini y Phi-4 Multimodal.

Sí, el rendimiento de un proveedor puede variar con el tiempo debido a cambios de infraestructura, balanceo de carga y actualizaciones. Medimos continuamente a todos los proveedores y mostramos las tendencias históricas de rendimiento en los gráficos "a lo largo del tiempo".

Al elegir un modelo en Microsoft Azure, considera: inteligencia (para tareas sensibles a la calidad), velocidad de salida (para tareas intensivas en rendimiento), latencia (para aplicaciones interactivas que requieren primeras respuestas rápidas), precios (para cargas de trabajo sensibles al costo) y funcionalidades como el tamaño de la ventana de contexto, el modo JSON o las llamadas a funciones.

Microsoft Azure: inteligencia, rendimiento y precio de sus modelos

Más inteligente

Más rápido

Menor precio

Inteligencia

Velocidad

Precio