Microsoft logo

MAI-Code-1.1-Flash

Multimodal
Microsoft

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. It is served by github-copilot with a 250K-token context window at $0.2 / $1.2 per 1M input/output tokens.

Key Specifications

Parameters
138.0B
Context
256.0K
Release Date
August 11, 2026
Average Score
67.8%

Timeline

Key dates in the model's history
Announcement
August 11, 2026
Last Update
September 12, 2026
Today
September 19, 2026

Technical Specifications

Parameters
138.0B
Training Tokens
-
Knowledge Cutoff
-
Family
-
Capabilities
MultimodalZeroEval

Pricing & Availability

Input (per 1M tokens)
$0.20
Output (per 1M tokens)
$1.20
Max Input Tokens
256.0K
Max Output Tokens
-
Supported Features
Function CallingStructured OutputCode ExecutionWeb SearchBatch InferenceFine-tuning

Benchmark Results

Model performance metrics across various tests and benchmarks

Programming

Programming skills tests
SWE-Bench Verified
GitHub Copilot production harness (VS Code-based)Self-reported
72.6%

Other Tests

Specialized benchmarks
Terminal-Bench 2.1
GitHub Copilot production harness (VS Code-based)Self-reported
62.9%

License & Metadata

License
proprietary
Announcement Date
August 11, 2026
Last Updated
September 12, 2026

Compare MAI-Code-1.1-Flash

All comparisons

Similar Models

All Models

Recommendations are based on similarity of characteristics: developer organization, multimodality, parameter size, and benchmark performance. Choose a model to compare or go to the full catalog to browse all available AI models.