Iniciar sesión
MAI-Code-1-Flash logo

MAI-Code-1-Flash

Microsoft · Lanzado jun 2026

Ver preciosComparar modelos
Condicional

MAI-Code-1-Flash was Microsoft's first in-house coding model, a sparse MoE priced at $0.75/$4.50 per 1M tokens. GitHub retired it from Copilot on 10 September 2026 and names MAI-Code-1.1-Flash, at $0.20/$1.20, as the replacement; it no longer appears in Copilot's supported-models or pricing tables. Copilot users should switch to 1.1-Flash, which is cheaper and scores higher on Terminal-Bench 2.1.

¿Es adecuada para ti?

Bueno para

  • Historical comparison only: GitHub retired it from Copilot on 10 September 2026, so new Copilot work should use MAI-Code-1.1-Flash
  • Fast, iterative agentic coding workflows where latency and token cost matter more than frontier reasoning, on any host that still serves it

No recomendado para

  • New Copilot workflows: the model no longer appears in GitHub Copilot's supported-models or pricing tables after its 10 September 2026 retirement
  • Cost-sensitive teams: its successor MAI-Code-1.1-Flash lists at $0.20/$1.20 per 1M against this model's $0.75/$4.50
  • Complex multi-file architectural changes or novel algorithm design requiring frontier-level code reasoning
  • General-purpose non-coding tasks — this is a specialized coding model, not a general assistant

Rendimiento por tarea

Autocompletado de código

Excellent

Excelente — baja latencia, longitud de respuesta adaptativa, diseñado específicamente para el bucle interactivo de Copilot

Refactorización de código

Very Good

Sólido — maneja bien las refactorizaciones rutinarias; un 60% menos de tokens que la competencia en SWE-Bench Verified

Seguimiento de instrucciones (codificación)

Excellent

El mejor de su categoría ligera — +28.9 puntos en IF Bench sobre Haiku 4.5; 85.8% en razonamiento adversarial

Uso agentivo de herramientas (entorno Copilot)

Very Good

Sólido — entrenado con el entorno de producción de Copilot; supera a Haiku 4.5 en τ¹-Bench

Corrección de errores reales (SWE-Bench Pro)

Good

Correcto para su tamaño — 51.2% de aciertos; 16 puntos de ventaja sobre Haiku 4.5 pero muy por debajo de los modelos frontera

Arquitectura multiarchivo / diseño novedoso

Fair

Débil — los 5B de parámetros activos limitan la profundidad de razonamiento; usa un modelo insignia para trabajo complejo

Razonamiento matemático y científico

Very Good

Bueno — 92.5% en AIME 2026, 84.6% en GPQA Diamond; sólido para un modelo ligero

Chat de propósito general / no programación

Poor

No está diseñado para ello — es un modelo de código especializado; usa GPT-5.5 o Claude para tareas generales

Precios

Entrada

$0.75 / 1M tokens

Salida

$4.50 / 1M tokens

Contexto

Retired from Copilot Sep 10 2026; 256K ctx

Ver precios completos

Pruebas de rendimiento

PruebaPuntuaciónFuente
SWE-Bench Pro51.2% (vs Haiku 4.5: 35.2%) Fuente
SWE-Bench VerifiedBeats Haiku 4.5 with up to 60% fewer tokens Fuente
AIME 202692.5% (vs Haiku 4.5: 83.3%) Fuente
GPQA Diamond84.6% (vs Haiku 4.5: 73.2%) Fuente
IF Bench75.0 (vs Haiku 4.5: 46.1) Fuente
Adversarial reasoning85.8% adjusted accuracy Fuente
Frontier Math (Tier 1-3)6.3% (vs Haiku 4.5: 2.8%) Fuente
Frontier Science58.2% (vs Haiku 4.5: 42.3%) Fuente
HLE18.0% (vs Haiku 4.5: 9.5%) Fuente

Historial de veredictos

26 sept 2026
conditional -> conditional (summary corrected: retired from Copilot): rating_before=conditional, rating_after=conditional. The summary still called it the model for high-volume Copilot workflows. GitHub's changelog set its deprecation for 2026-09-10, and on 2026-09-26 it is absent from Copilot's supported-models and pricing tables. The rating is held pending an operator ruling, since a retired model may merit caution or avoid.
15 jul 2026
copy edit to brevity standard: detail preserved in existing structured children

Fuentes

Registro de verificación

  • Precios· Sin cambios

    Agente automatizado

  • Precios· Sin cambios

    Agente automatizado

  • Estado· Actualizado

    Agente automatizado

    GitHub's changelog set MAI-Code-1-Flash's deprecation for 2026-09-10 and names MAI-Code-1.1-Flash as the replacement. On 2026-09-26 the model is absent from both the Copilot models-and-pricing table and the supported-models table, while MAI-Code-1.1-Flash is listed GA at $0.20/$1.20. Fixed via model_update: pricing_context is now 'Retired from Copilot Sep 10 2026; 256K ctx'; strengths and audiences were rewritten to drop the 'all Copilot tiers' and 'stretch Copilot credits' claims and to point to 1.1-Flash; the changelog and both GitHub docs pages were added as sources. The summary was corrected via verdict_set e9dd3180, rating_before=conditional and rating_after=conditional; the rating is held pending an operator ruling. Price scalars are unchanged. No draft post: the retirement was announced on 2026-08-11 and took effect 16 days ago.

  • Precios· Sin cambios

    Agente automatizado

  • Precios· Sin cambios

    Agente automatizado

  • Precios· Sin cambios

    Agente automatizado

  • Precios· Sin cambios

    Agente automatizado

  • Precios· Sin cambios

    Agente automatizado

  • Precios· Sin cambios

    Agente automatizado

    Pricing unchanged: $0.75/$4.50 per 1M tokens, 256K context. Confirmed live on GitHub Copilot pricing page.

  • Precios· Sin cambios

    Importado en el lanzamiento

  • Perfil· Sin cambios

    Importado en el lanzamiento

Cómo evaluamos