MAI Code 2 Flash
Microsoft coding model launched November 17, 2026 for code explanation, targeted edits, routine fixes, and fast iteration, with image input and a 1.05 million-token context window.
At a glance
- Context window
- 1.05M tokens
- Max output
- 128K tokens
- Text input / output · per 1M tokens
- $0.050 / $0.250
- Knowledge cutoff
- —
Benchmark snapshot
Best result in each category
CodingTerminal-Bench 2.1
78.3Max (With Tools) · #42 of 68
Benchmark details
2 results · Best published score per benchmark, grouped by category
| Benchmark · setting | Score | Rank |
|---|---|---|
| Coding | ||
| DeepSWE Max (With Tools) · Repository Engineering | 59.9 | 48 / 73 |
| Terminal-Bench 2.1 Max (With Tools) · Agentic Development | 78.3 | 42 / 68 |
Rank is this model's position among every model with a published result on that leaderboard; the bar shows how far up the field that is. Orange marks first place on leaderboards with at least 5 models. Unranked leaderboards publish a composite or direction-less score.
Back to benchmark snapshot ↑API pricing
Current promotional text rates, USD per million tokens · standard list effective 2027-05-17
- Input
- $0.05
- Output
- $0.25
- Cached input
- $0.005
Blended price4:1 input:output · 90% of input tokens cached · same basis as the price ranking
$0.0576/ 1M tokens
Promotions & time-of-day pricing
The prices below apply only within the stated dates and times. Standard prices are used outside these windows.
GitHub Copilot launch offer · standard · input
$0.10$0.05 / 1,000,000 tokens50% offValid 2026-11-17 – 2027-05-16 · UTC
GitHub Copilot launch offer · standard · Cache read
$0.01$0.005 / 1,000,000 tokens50% offValid 2026-11-17 – 2027-05-16 · UTC
GitHub Copilot launch offer · standard · output
$0.50$0.25 / 1,000,000 tokens50% offValid 2026-11-17 – 2027-05-16 · UTC
Specifications
- Provider
- Microsoft
- Model type
- Reasoning model
- Released
- November 17, 2026
- Total parameters
- 398B
- Active parameters
- 12B
- Architecture
- Mixture of experts
- Context window
- 1.05M tokens
- Max output
- 128K tokens
- Input
- Text, Image
- Output
- Text
- Weights
- Closed / API only
- Commercial use
- Proprietary
Capabilities
Model analysis
Microsoft coding model launched November 17, 2026 for code explanation, targeted edits, routine fixes, and fast iteration, with image input and a 1.05 million-token context window. Microsoft trained the family from scratch and ran reinforcement learning in GitHub Copilot CLI and VS Code agent harnesses.
The two models share an architecture and differ in scale
The sparse mixture-of-experts decoder has an integrated vision encoder and about 398 billion total and 12 billion active parameters. Microsoft reports roughly 25 trillion pretraining tokens, including 22 trillion text and 3 trillion image tokens, with logit distillation from an internal teacher, followed by mid-training and context extension. Both models accept up to 922,000 input tokens and return up to 128,000 text tokens. Reasoning effort runs from low through medium, high, xhigh, and max. At launch, Microsoft optimized the family for English-language code production.
The reported pass rates depend on the test setup
Microsoft reports 59.9% / 78.3% / 44.8% on DeepSWE 1.1, Terminal-Bench 2.1, and SWE Atlas respectively. It used MSBench and GitHub Copilot CLI 1.0.102 with max reasoning and subagents disabled, averaging single-attempt pass rates over three runs per task. Microsoft reports comparable performance to GPT-6 Luna and Claude Haiku 5.5 across the three tests, at about 65% and 20% of their costs respectively. Microsoft used launch promotional rates for its cost comparisons. In a separate advanced-autonomy evaluation, the card reports 56.36% / 27.4% / 65.2% on GAIA Validation, DeepSWE 1.1, and SWE Atlas respectively.
Microsoft reports average costs of $0.098, $0.023, and $0.034 per task on these three tests at launch rates, respectively.
GitHub Copilot launch pricing runs through May 2027
GitHub Copilot lists Flash as GA for Free, Student, Pro, Pro+, Max, Business, and Enterprise. Microsoft Foundry offers Flash in limited-access preview via an OpenAI-compatible Responses API. Copilot charges $0.05 / $0.005 / $0.25 per million input, cached-input, and output tokens respectively. From May 17, 2027 the rates are $0.10 / $0.01 / $0.50. The models are hosted, with no public weight downloads.
Microsoft also announced planned availability through OpenRouter, Vercel, Hermes, OpenCode, and Kilo.
Sources
Last reviewed 2026-11-18
MAI Code 2 Flash vs competitors
Comparable results only · record is MAI Code 2 Flash's wins–losses
| Benchmark | MAI Code 2 FlashThis model | vs MAI Code 2Record 0–2 | vs Inkling 2 SmallRecord 0–2 | vs SWE-2 FlashRecord 0–2 | vs DeepSeek-V4.1-Flash-1017Record 0–2 |
|---|---|---|---|---|---|
| DeepSWE | 59.9 | 73.2 | 72.1 | 60.1 | 77.0 |
| Terminal-Bench 2.1 | 78.3 | 84.6 | 85.1 | 89.2 | 91.0 |
Primary sources
- MAI Code 2 Flash technical reportmicrosoft.ai
- MAI Code 2 Flash online demogithub.com
Community discussion
Share the API, setup, and tradeoffs you actually tested. Rate the model once; comment as often as you like.
Used this model? Sign in with Google to rate it and share what worked, what did not, and which API or setup you tested.
No community comments yet.
