Plateforme d'intelligence des modèles IA
← Retour à tous les modèles

Mistral Small 4

Mistral

Fournisseur
Mistral
Type
chat
Fenêtre de contexte
128 000
Prix d'entrée
$0.15/ 1M tokens
Prix de sortie
$0.60/ 1M tokens
Date de sortie
2026-02-25

Data Trust Card

  • Model verifiedExperimental
  • Pricing verifiedExperimental
  • Benchmark verifiedExperimental

FreshnessNeeds verification

Confidence: 50%

Evidence Timeline

Price updated: {date}
2026-08-11

Related Models

Similar Models

  • Qwen3 235BConfidence 57%

    Similar coding, function_calling, long_context with comparable pricing (sim 63.7%)

  • Hunyuan TurboSConfidence 57%

    Similar coding, function_calling, long_context with comparable pricing (sim 62.3%)

  • Qwen3 32BConfidence 56%

    Similar coding, function_calling, long_context with comparable pricing (sim 61.2%)

  • Doubao 1.5 ProConfidence 56%

    Similar coding, function_calling, long_context with comparable pricing (sim 60.4%)

Lower cost alternatives

  • Qwen3 235B

    Lower input price ($0.15/1M vs $0.2/1M) with coding, function_calling capabilities

  • Qwen3 Coder 480B

    Lower input price ($0.15/1M vs $0.3/1M) with coding, function_calling capabilities

  • Claude 3.5 Sonnet

    Lower input price ($0.15/1M vs $3/1M) with coding, function_calling capabilities

Higher performance alternatives

  • DeepSeek Reasoner

    Alternative choice with similar benchmark performance and coding, function_calling capabilities

  • GPT-4.1 nano

    Alternative choice with similar benchmark performance and coding, function_calling capabilities

  • ERNIE 4.5 VL 424B

    Alternative choice with similar benchmark performance and coding, function_calling capabilities

Engine-computed relationship (trust-gated)

Description

Mistral's fast, affordable model for high-volume and lightweight tasks.

Cas d'utilisation

  • high-volume
  • classification
  • chatbot
  • extraction

Statut des données

Statut
Actif
Dernière vérification
Non vérifié

Classement général

Global

18.9

Benchmark

0

Capacité

57.1

Efficacité prix

100

Contexte

64

Global = Benchmark 50% + Capacité 20% + Efficacité prix 20% + Contexte 10%.

Frequently Asked Questions

What is Mistral Small 4?

Mistral Small 4 is an AI model by Mistral. Mistral's fast, affordable model for high-volume and lightweight tasks.

What is Mistral Small 4's context window?

Mistral Small 4 supports a context window of 128 000 tokens.

How much does Mistral Small 4 cost?

Mistral Small 4 costs $0.15 per 1M input tokens and $0.60 per 1M output tokens.

What can Mistral Small 4 do?

Mistral Small 4 supports: Raisonnement, Codage, Appel d'outils, Contexte long.

Related Resources