MAI-Code-1.1-Flash

Fiche modèle · Microsoft · mai-code-1.1-flash

MAI-Code-1.1-Flash is a Microsoft text-to-text and image-to-text coding model built for fast, efficient assistance in everyday developer workflows, including agentic coding in real repositories, repository Q&A, refactoring, and tool-using developer scenarios in GitHub Copilot. Building on MAI-Code-1-Flash, it adds native vision support and, per Microsoft, produces higher-quality code at about 25% greater token efficiency and roughly one quarter of the cost of the June MAI-Code-1-Flash launch. It is GA in GitHub Copilot; for annual Copilot subscribers on request-based billing, GitHub charges it at a 0.25× premium request multiplier.

Identité

OrganisationMicrosoft
Pays—
Famille—
Typellm
Date de sortie2026-08-11
Licenceproprietary
Licence commercialenon
Licence open sourcenon
Poids ouvertsnon
Auto-hébergement (DGX Spark)non — poids fermés
Modalitésimage|text
Fenêtre de contexte—
KV cache / token (Q8)—
Paramètres totaux138 B
Paramètres actifs (session)5 B
Source des paramètres actifsweb : https://microsoft.ai/pdf/MAI-Code-1.1-Flash-Model-Card.PDF
Curation des paramètres actifs2026-10-19
Prix par session0.44 $ (1 000 000 tokens entrée + 200 000 sortie, prix le moins cher par fournisseur)
Tokens / tâche (Data Players)1.2 M
Tokens / tâche (Terminal-Bench)83.3 M (est.)
Ratio Terminal-Bench / Data Players69.4×
Coût / tâche (Terminal-Bench)3.70 $ (est.)
Tokens d'entraînement—
Coupure de connaissance—
Créé le2026-08-13T17:05:20.687213Z
Mis à jour le2026-11-04T06:04:52.870827Z
SourceLLM Stats
URLhttps://llm-stats.com/models/mai-code-1.1-flash

Prix par fournisseur

Statut du modèle (offres + auto-hébergement) : non.

Toutes sources confondues : catalogue llm-stats, routeurs EUrouter/Cortecs et catalogues directs (OVHcloud, Scaleway, Infomaniak). Les prix sont convertis en USD pour la comparaison ; le prix publié et le taux BCE utilisé sont dans l'infobulle. Les prix inconnus restent indiqués « — ».

FournisseurSourceSouverainetéEntrée (/M tok)Sortie (/M tok)QuantificationContexteLatence (s)Débit (tok/s)Statut
GitHub Copilotllm-statsnon0.20 $1.20 $————active

Souveraineté qualifiée par offre (modèle × fournisseur) ; survolez le badge pour les critères.

Souveraineté

Scores par catégorie (top_scores)

CatégorieScore
reasoning0.70
frontend_development0.70
general0.70
agents0.60
code0.70
tool_calling0.60

Scores par benchmark

BenchmarkCatégorieScoreMaxNormalisé (API)VérifiéAuto-déclaréDaté du
swe-bench-verifiedreasoning0.731.000.73nonoui2026-11-04T06:04:52.878658Z
terminal-bench-2.1reasoning0.631.000.63nonoui2026-11-04T06:04:52.883062Z

Classements

CatégorieRangScoreNote conservatriceBenchmarks évaluésMéthodeDaté duProvenance
agents128—10.361trueskill2026-10-19model_history
code123—18.562trueskill2026-10-19model_history

Lecture : Note conservatrice = borne basse TrueSkill (niveau estimé − marge d’incertitude) ; Benchmarks évalués = nombre de benchmarks de la catégorie pris en compte. À performances égales, un modèle évalué sur moins de benchmarks obtient une note plus basse : la note ne se déduit pas d’un seul score. Provenance : rankings = top 50 de /rankings ; model_history = historique du modèle (/models/{id}/history), utilisé pour les rangs au-delà du top 50.

Données

Dates de curation (récupération depuis l'API) : models 2026-11-04 · benchmarks 2026-11-04 · scores 2026-11-04 · updates 2026-11-04 · account 2026-11-04 · model_history 2026-11-04 · rankings 2026-11-04