Claude Opus 4.8 — Especificaciones
Compiled by Mustafa Ihsan from the vendor’s published documentation · Last updated
| Desarrollador | Anthropic |
|---|---|
| Tipo | LLM (razonamiento) |
| Modalidad | Texto y visión → texto |
| Parámetros | No revelado |
| Ventana de contexto | 1 millón |
| Salida máxima | 128 K |
| Licencia | Propietario |
| Pesos abiertos | No |
| Lanzado | 2026 |
| Precio de entrada | $5.00 /1M |
| Precio de salida | $25.00 /1M |
| Proveedores de API | Anthropic, AWS, Vertex AI, Azure |
What is Claude Opus 4.8?
Claude Opus 4.8 is Anthropic’s Opus-tier flagship for state-of-the-art long-horizon
agentic work, coding and knowledge tasks. It runs adaptive thinking only, and carries a
1M-token context window at standard pricing — there is no long-context premium, which
distinguishes it from several rivals that step the input rate up once a prompt passes a
threshold. At $5 in / $25 out per million tokens it sits squarely in the frontier bracket.
The no-premium 1M context is the practical reason to choose it. Models that double their
input rate above 200K tokens make large-context work unpredictable to budget: the same
feature costs a different amount depending on how much a user pasted in. Opus 4.8 bills the
same rate from the first token to the millionth, so a retrieval-heavy application can size
its prompts around what produces the best answer rather than around a pricing cliff. It has
since been joined by Claude Opus 5, which ranks higher on the Artificial Analysis
Intelligence Index at identical pricing — so for new builds, check whether Opus 5 is
available in your region and provider before defaulting to 4.8.
Claude Opus 4.8 pricing: API cost per 1M tokens
| Entrada (por cada millón de tokens) | $5.00 |
|---|---|
| Salida (por cada millón de tokens) | $25.00 |
| Relación salida/entrada | 5× |
| Combinada (4:1 entrada:salida) | $9.00 por 1 millón de tokens |
What Claude Opus 4.8 costs per month
Gasto mensual real con una mezcla entrada:salida de 4:1, es decir, la proporción que realmente genera una carga de trabajo típica de chat o RAG.
| Carga de trabajo | Tokens/mes | Coste mensual |
|---|---|---|
| Proyecto secundario | 1 millón de tokens de entrada / 0,25 millones de tokens de salida | $11 |
| Pequeño equipo | 20 millones de tokens de entrada / 5 millones de tokens de salida | $225 |
| Producción | 200 millones de tokens de entrada / 50 millones de tokens de salida | $2,250 |
Calcule sus propios números en la Calculadora de costos de API de IA.
Cheaper alternatives to Claude Opus 4.8
| Modelo | Dólares por millón combinados | Usted ahorra |
|---|---|---|
| Kimi K3 abierta | $5.40 | 40 % más barato |
| GLM 5.2 abierta | $2.00 | 78 % más barato |
| Gemini 3.5 Flash | $3.00 | un 67 % más barato |
Preguntas frecuentes
How much does Claude Opus 4.8 cost per 1M tokens?
Claude Opus 4.8 costs $5.00 per 1M input tokens and $25.00 per 1M output tokens. At a typical 4:1 input-to-output mix that blends to about $9.00 per 1M tokens.
How much does Claude Opus 4.8 cost per month?
A small-team workload of 20M input and 5M output tokens a month costs about $225 on Claude Opus 4.8. A side project (1M in / 0.25M out) costs roughly $11.25.
What is a cheaper alternative to Claude Opus 4.8?
Kimi K3 is the strongest cheaper option in our database at $5.40 per 1M blended — about 40% less than Claude Opus 4.8. It is also open-weight, so self-hosting is an option.
Can I run Claude Opus 4.8 locally?
No. Claude Opus 4.8 is a closed, API-only model — the weights are not released, so it cannot be self-hosted.
Why does Claude Opus 4.8 charge more for output than input?
Output tokens are generated one at a time and cannot be batched the way a prompt can, so they cost the provider more to serve. Claude Opus 4.8 charges 5× more for output, which is why prompt-heavy workloads are far cheaper to run than generation-heavy ones.
Vea todos los modelos Claude comparados lado a lado: Precios de la API de Claude.
Los precios corresponden a las tarifas oficiales publicadas para la API principal del modelo y se revisan periódicamente conforme los proveedores los actualicen. No incluyen descuentos por volumen, procesamiento por lotes ni entradas en caché. Compare todos los modelos uno al lado del otro en la Base de datos de modelos de IA o el Clasificación de modelos de lenguaje grande (LLM).