CoreWeave: inteligencia, rendimiento y precio de sus modelos

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

CoreWeave

Análisis de los modelos de CoreWeave 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 CoreWeave para tu caso de uso.

Más inteligente

GLM-5.2 (max)

MiniMax-M3

Kimi K2.6

Kimi K2.7 Code

GLM-5.1

Índice de inteligencia

22 modelos en total

Más rápido

Kimi K2.7 Code

288 t/s

MiniMax-M3

281 t/s

Kimi K2.6

268 t/s

Kimi K2.6

234 t/s

Nemotron 3 Ultra

200 t/s

Velocidad de salida

22 modelos en total

Menor precio

gpt-oss-20b (high)

$0.04

gpt-oss-20b (low)

$0.04

gpt-oss-120b (high)

$0.05

gpt-oss-120b (low)

$0.05

Granite 4.1 8B

$0.06

Precio combinado (por 1M de tokens)

22 modelos en total

Indica un modelo de razonamiento

CoreWeave ofrece 22 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 CoreWeave son GLM-5.2 (max) (51), MiniMax-M3 (44) y Kimi K2.6 (44).
En velocidad de salida, los modelos más rápidos son Kimi K2.7 Code (288 t/s), MiniMax-M3 (281 t/s) y Kimi K2.6 (268 t/s).
En latencia, Granite 4.1 8B (0.80s), Llama 3.3 70B (0.93s) y Llama 3.1 8B (1.02s) ofrecen el menor tiempo hasta el primer token de respuesta.
En precios, gpt-oss-20b (high) ($0.04), gpt-oss-20b (low) ($0.04) y gpt-oss-120b (high) ($0.05) ofrecen los menores precios combinados por 1M de tokens.
En tamaño de ventana de contexto, DeepSeek V4 Flash (high) (1M), DeepSeek V4 Flash (1M) y GLM-5.2 (max) (262k) admiten las ventanas de contexto más grandes en CoreWeave.

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
GLM-5.2 (max)	262k	Abierto	51	--	132	1.33	20.34	15.21	Modelo Proveedor
MiniMax-M3	262k	Abierto	44	$0.12	281	1.18	10.09	7.12	Modelo Proveedor
Kimi K2.6	262k	Abierto	44	--	268	1.18	19.64	16.60	Modelo Proveedor
Kimi K2.7 Code	262k	Abierto	42	$0.31	288	1.19	10.67	7.74	Modelo Proveedor
GLM-5.1	203k	Abierto	40	$0.49	121	1.25	36.77	31.38	Modelo Proveedor
GLM-5 (FP8)	200k	Abierto	40*	--	--	--	--	--	Modelo Proveedor
Nemotron 3 Ultra	262k	Abierto	38	$0.15	200	1.08	14.94	11.36	Modelo Proveedor
DeepSeek V4 Flash (high)	1M	Abierto	37	$0.07	33	1.60	54.30	37.55	Modelo Proveedor
Kimi K2.6	262k	Abierto	35*	--	234	1.17	3.31	--	Modelo Proveedor
MiniMax-M2.5	197k	Abierto	34*	--	83	1.10	31.25	24.12	Modelo Proveedor
Gemma 4 31B	262k	Abierto	29	--	34	1.02	66.41	50.77	Modelo Proveedor
DeepSeek V4 Flash	1M	Abierto	29*	--	38	2.27	15.49	--	Modelo Proveedor
NVIDIA Nemotron 3 Super	262k	Abierto	25	$0.20	142	1.00	18.58	14.06	Modelo Proveedor
gpt-oss-120b (high)	131k	Abierto	24	$0.02	31	1.47	81.27	63.83	Modelo Proveedor
DeepSeek V3.1	128k	Abierto	21*	--	65	1.41	9.05	--	Modelo Proveedor
Qwen3 Coder 480B	262k	Abierto	18*	--	65	1.20	8.85	--	Modelo Proveedor
gpt-oss-120b (low)	131k	Abierto	15	$0.0045	30	1.51	85.87	67.49	Modelo Proveedor
gpt-oss-20b (high)	131k	Abierto	15	$0.01	175	0.90	15.22	11.45	Modelo Proveedor
gpt-oss-20b (low)	131k	Abierto	14*	--	164	0.90	16.17	12.22	Modelo Proveedor
Llama 3.3 70B	128k	Abierto	9	$0.10	75	0.93	7.63	--	Modelo Proveedor
Qwen3 30B A3B 2507	262k	Abierto	9*	--	64	1.05	8.83	--	Modelo Proveedor
Llama 3.1 8B	128k	Abierto	8	--	139	1.02	4.60	--	Modelo Proveedor
Granite 4.1 8B	131k	Abierto	7*	--	94	0.80	6.12	--	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 CoreWeave

CoreWeave ofrece 22 modelos que monitoreamos: GLM-5.2 (max), MiniMax-M3, Kimi K2.6, Kimi K2.7 Code, GLM-5.1, Nemotron 3 Ultra, DeepSeek V4 Flash (high), Kimi K2.6, MiniMax-M2.5, Gemma 4 31B, DeepSeek V4 Flash, NVIDIA Nemotron 3 Super, gpt-oss-120b (high), DeepSeek V3.1, Qwen3 Coder 480B, gpt-oss-120b (low), gpt-oss-20b (high), gpt-oss-20b (low), Llama 3.3 70B, Qwen3 30B A3B 2507, Llama 3.1 8B y Granite 4.1 8B.

El modelo más inteligente disponible en CoreWeave es GLM-5.2 (max), con una puntuación de 51 en el Índice de Inteligencia.

El modelo más rápido en CoreWeave por velocidad de salida es Kimi K2.7 Code, con 287.9 tokens por segundo.

El modelo con el menor tiempo hasta el primer token de respuesta en CoreWeave es Granite 4.1 8B, con 0.80s. Una menor latencia significa una respuesta inicial más rápida.

El modelo más económico en CoreWeave por precio combinado es gpt-oss-20b (high), a $0.04 por 1M de tokens (proporción 7:2:1 de aciertos de caché/entrada/salida).

Los precios en CoreWeave varían hasta 43x entre modelos, desde $0.04 por 1M de tokens para gpt-oss-20b (high) hasta $1.70 por 1M de tokens para GLM-5.1.

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

Sí, los 22 modelos en CoreWeave admiten el modo JSON para salida estructurada.

Sí, los 22 modelos en CoreWeave admiten llamadas a funciones (uso de herramientas).

Sí, CoreWeave ofrece 14 modelos de razonamiento: GLM-5.2 (max), MiniMax-M3, Kimi K2.6, Kimi K2.7 Code, GLM-5.1, Nemotron 3 Ultra, DeepSeek V4 Flash (high), MiniMax-M2.5, Gemma 4 31B, NVIDIA Nemotron 3 Super, gpt-oss-120b (high), gpt-oss-120b (low), gpt-oss-20b (high) y gpt-oss-20b (low). Los modelos de razonamiento usan pensamiento extendido para resolver problemas complejos antes de responder.

Sí, los 22 modelos en CoreWeave son modelos de pesos abiertos.

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 CoreWeave, 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.

CoreWeave: inteligencia, rendimiento y precio de sus modelos

Más inteligente

Más rápido

Menor precio

Inteligencia

Velocidad

Precio