Parasail: inteligencia, rendimiento y precio de sus modelos

Parasail
Parasail

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

Más inteligente

#1
Kimi K3
Kimi K3
57
#2
GLM-5.2 (max) (NVFP4)
GLM-5.2 (max) (NVFP4)
51
#3
MiniMax-M3 (MXFP8)
MiniMax-M3 (MXFP8)
44
#4
Kimi K2.6
Kimi K2.6
44
#5
Kimi K2.7 Code
Kimi K2.7 Code
42

Índice de inteligencia

30 modelos en total

Más rápido

#1
GLM-5.2 (max) (NVFP4)
GLM-5.2 (max) (NVFP4)
203 t/s
#2
gpt-oss-120b (high)
gpt-oss-120b (high)
192 t/s
#3
Trinity Large Thinking (FP8)
Trinity Large Thinking (FP8)
186 t/s
#4
gpt-oss-120b (low)
gpt-oss-120b (low)
185 t/s
#5
Qwen3 Next 80B A3B
Qwen3 Next 80B A3B
140 t/s

Velocidad de salida

30 modelos en total

Menor precio

#1
MiMo-V2.5
MiMo-V2.5
$0.09
#2
Gemma 4 26B A4B
Gemma 4 26B A4B
$0.10
#3
Gemma 4 26B A4B
Gemma 4 26B A4B
$0.10
#4
DeepSeek V4 Flash (max) (FP8)
DeepSeek V4 Flash (max) (FP8)
$0.11
#5
DeepSeek V4 Flash (high) (FP8)
DeepSeek V4 Flash (high) (FP8)
$0.11

Precio combinado (por 1M de tokens)

30 modelos en total

Indica un modelo de razonamiento

Parasail ofrece 30 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 Parasail son Kimi K3 (57), GLM-5.2 (max) (NVFP4) (51) y MiniMax-M3 (MXFP8) (44).
  • En velocidad de salida, los modelos más rápidos son GLM-5.2 (max) (NVFP4) (203 t/s), gpt-oss-120b (high) (192 t/s) y Trinity Large Thinking (FP8) (186 t/s).
  • En latencia, Llama 4 Maverick (FP8) (0.84s), Qwen3 Next 80B A3B (0.96s) y Qwen3 Coder Next (FP8) (1.13s) ofrecen el menor tiempo hasta el primer token de respuesta.
  • En precios, MiMo-V2.5 ($0.09), Gemma 4 26B A4B ($0.10) y Gemma 4 26B A4B ($0.10) ofrecen los menores precios combinados por 1M de tokens.
  • En tamaño de ventana de contexto, Kimi K3 (1M), DeepSeek V4 Flash (max) (FP8) (1M) y DeepSeek V4 Flash (high) (FP8) (1M) admiten las ventanas de contexto más grandes en Parasail.

Aspectos destacados

Artificial Analysis Intelligence Index · Higher is better
Output tokens per second · Higher is better
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

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

Agentic tool use

Agentic coding & terminal use

Coding

Reasoning & knowledge

Scientific reasoning

Physics reasoning

Long context reasoning

Agentic knowledge work, Elo

Agentic SaaS workflows

Legal agentic work, criterion pass rate

Agentic business operations

Instruction following

Long-horizon agentic tasks

Kubernetes incident root-cause analysis

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 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 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.

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.

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.

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.

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
Logo de Kimi
Kimi K3
1.05M
Abierto
57
$0.68
68
1.97
38.81
29.47
Logo de Z AI
GLM-5.2 (max) (NVFP4)
1M
Abierto
51
--
203
1.24
13.58
9.87
Logo de MiniMax
MiniMax-M3 (MXFP8)
1M
Abierto
44
$0.16
92
1.27
28.38
21.69
Logo de Kimi
Kimi K2.6
262k
Abierto
44
$0.29
73
1.55
69.07
60.70
Logo de Kimi
Kimi K2.7 Code
262k
Abierto
42
$0.16
67
1.68
42.39
33.26
Logo de DeepSeek
DeepSeek V4 Flash (max) (FP8)
1.05M
Abierto
40
--
77
2.04
81.19
72.67
Logo de Z AI
GLM-5.1 (FP8)
203k
Abierto
40
$0.21
77
2.41
58.02
49.13
Logo de Z AI
GLM-5 (FP8)
203k
Abierto
40*
--
39
1.64
93.55
79.16
Logo de DeepSeek
DeepSeek V4 Flash (high) (FP8)
1.05M
Abierto
37
--
85
1.94
22.36
14.55
Logo de Xiaomi
MiMo-V2.5
1M
Abierto
37
$0.03
76
1.14
33.87
26.18
Logo de Z AI
GLM-5.1
203k
Abierto
35*
--
78
2.38
8.79
--
Logo de Kimi
Kimi K2.6 (INT4)
262k
Abierto
35*
--
73
1.50
8.30
--
Logo de Alibaba
Qwen3.5 397B A17B
262k
Abierto
34
$0.22
71
1.04
52.90
44.82
Logo de MiniMax
MiniMax-M2.5 (FP8)
197k
Abierto
34*
--
80
1.27
32.55
25.02
Logo de Alibaba
Qwen3.6 35B A3B
262k
Abierto
32
$0.19
81
1.33
74.25
66.74
Logo de Google
Gemma 4 31B
262k
Abierto
29
$0.02
44
3.45
54.68
39.78
Logo de Google
Gemma 4 26B A4B
256k
Abierto
26
$0.02
44
1.92
59.20
45.82
Logo de Alibaba
Qwen3.6 35B A3B (FP8)
262k
Abierto
24
$0.41
87
1.27
7.02
--
Logo de OpenAI
gpt-oss-120b (high)
131k
Abierto
24
--
192
0.77
13.80
10.42
Logo de Google
Gemma 4 31B
262k
Abierto
22
$0.03
32
2.81
18.40
--
Logo de Alibaba
Qwen3 Coder Next (FP8)
262k
Abierto
21
$0.11
85
1.13
7.03
--
Logo de Google
Gemma 4 26B A4B
262k
Abierto
20*
--
32
2.09
17.90
--
Logo de Arcee AI
Trinity Large Thinking (FP8)
262k
Abierto
18
$0.07
186
0.96
14.39
10.75
Logo de Alibaba
Qwen3 235B 2507
262k
Abierto
18*
--
31
1.36
17.61
--
Logo de OpenAI
gpt-oss-120b (low)
131k
Abierto
15
--
185
0.78
14.31
10.83
Logo de Alibaba
Qwen3 VL 235B A22B (FP8)
131k
Abierto
14*
--
35
1.38
15.63
--
Logo de Meta
Llama 4 Maverick (FP8)
1.05M
Abierto
14
$0.02
122
0.84
4.95
--
Logo de Alibaba
Qwen3 Next 80B A3B
262k
Abierto
14*
--
140
0.96
4.53
--
Logo de Meta
Llama 3.3 70B (FP8)
131k
Abierto
9
$0.02
67
2.74
10.24
--
Logo de Allen Institute for AI
Olmo 3.1 32B Think
65.5k
Abierto
8*
--
--
--
--
--
Logo de Google
Gemma 3 27B
131k
Abierto
7
$0.14
51
1.18
10.95
--
Logo de Allen Institute for AI
Olmo 3 7B
65.5k
Abierto
3*
--
--
--
--
--

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 Parasail

El modelo más inteligente disponible en Parasail es Kimi K3, con una puntuación de 57 en el Índice de Inteligencia.

El modelo más rápido en Parasail por velocidad de salida es GLM-5.2 (max) (NVFP4), con 202.6 tokens por segundo.

El modelo con el menor tiempo hasta el primer token de respuesta en Parasail es Llama 4 Maverick (FP8), con 0.84s. Una menor latencia significa una respuesta inicial más rápida.

El modelo más económico en Parasail por precio combinado es MiMo-V2.5, a $0.09 por 1M de tokens (proporción 7:2:1 de aciertos de caché/entrada/salida).

Los precios en Parasail varían hasta 25x entre modelos, desde $0.09 por 1M de tokens para MiMo-V2.5 hasta $2.31 por 1M de tokens para Kimi K3.

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

Sí, los 30 modelos en Parasail admiten el modo JSON para salida estructurada.

Sí, los 30 modelos en Parasail admiten llamadas a funciones (uso de herramientas).

Sí, los 30 modelos en Parasail 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 Parasail, 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.