Data as of November 17, 2026 · How the score is built
MAI Code 2
Decision readingMAI Code 2 is tracked, but not publicly ranked yet. The profile exposes 3 sourced benchmark rows and leaves unsupported fields blank until a published record exists.
Email me when its price, rank or availability changes.
Capability
Unranked
field median 49.5Not eligible for a public rank
Price
$1input / $5 output
input median $0.60cached $0.050 · blended $3
Speed
Not measured
field median 89 tok/sTime to first token not measured
Context
1.05Mtokens
field median 256,000Maximum output length is tracked separately
Decision snapshot
How much published evidence this model has, where that evidence is strongest, and what to validate before choosing it. It has no public rank, so no field position is shown.
Strongest published evidence
Published rows are visible, but no category has enough eligible evidence for a comparative rank.
Validate before choosing
3 published rows leave some tracked benchmark slots empty. Independent runtime speed has not been measured.
Source-linked · 3 displayable benchmark rows
Category score record
Scores and ranks appear only where published evidence can be displayed. The table keeps the score, weight, cohort, and evidence state together.
| Category | Score | Rank | Percentile | Weight | Benchmarks | Evidence |
|---|---|---|---|---|---|---|
| AgenticWeight 22%1 benchmarkVerified | Score pending | 1 benchmark | Verified | |||
| CodingWeight 20%2 benchmarksVerified | Score pending | 2 benchmarks | Verified | |||
| ReasoningWeight 17%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| MultimodalWeight 12%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| KnowledgeWeight 12%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| MultilingualWeight 7%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| Inst. FollowingWeight 5%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured | |||
| MathWeight 5%0 benchmarksNot measured | Not measured | 0 benchmarks | Not measured |
3 of 922 tracked benchmark slots have displayable evidence · bars run 0–100
Coverage detailsHow much of this is verified
Coverage is split by category so a strong number never hides a thin evidence base. Verified means the row is tied to a published source; provisional rows remain visible but separate.
- Agentic1/1 verified
- Coding2/2 verified
- ReasoningNot measured
- MultimodalNot measured
- KnowledgeNot measured
- MultilingualNot measured
- Inst. FollowingNot measured
- MathNot measured
Benchmark ledger
Every result says who ran the test. For this model, Microsoft ran all 4. Coding opens by default. The tick marks the best source-verified result in the catalog; provisional leaders do not set the reference.
Coding2 rows
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| DeepSWE | 73.2% | Best verified: Muse Wonder Full · 79.0% | Gap5.8 behind | Reported by Microsoft | Shown only |
| Terminal-Bench 2.1 | 84.6% | Best verified: SWE-2 · 92.8% | Gap8.2 behind | Reported by Microsoft | Shown only |
Agentic1 row
| Benchmark | Score | Versus best verified row | Gap | Who ran it | Use in ranking |
|---|---|---|---|---|---|
| Terminal-Bench 2.1 | 84.6% | Best verified: SWE-2 · 92.8% | Gap8.2 behind | Reported by Microsoft | Shown only |
Other lab-reported results1 rowAnother metric, split or setting, or a test without a page
| Benchmark | Score | On a 0–100 scale | Who ran it | Use in ranking |
|---|---|---|---|---|
| SWE AtlasPass rate · Microsoft's MSBench runs with the GitHub Copilot CLI harness (v1.0.102, no subagents), max reasoning effort, mean of three runs; 284 tasks across codebase Q&A, test writing and refactoring | 43.1% | Reported by Microsoft | Shown only |
Outlined bars are lab-reported. Filled bars are benchmark-run or independent. Bars run 0–100; the dark tick marks the best source-verified value.
Lineage
The sequence follows explicit supersedes links. Each score is estimated for that model; a relative can inform a sparse estimate but never sets a floor, so a newer release can score below an earlier one. Scores and prices remain blank when the corresponding public row or first-party rate is unavailable.
Spec sheet
Each documented value carries its source. Missing fields stay visible as not sourced or not published, rather than disappearing from the page.
- API model ID
- MAI Code 2GitHub Copilot models and pricing
- Context window
- 1.05MGitHub Copilot models and pricing
- Maximum output
- Not sourced yet
- Knowledge cutoff
- Not sourced yet
- Input modalities
- Not sourced yet
- Output modalities
- Not sourced yet
- Parameters
- Not disclosed by the provider
- Availability
- Available in GitHub Copilot since November 17, 2026 on the Pro, Pro+, Max, Business, and Enterprise plans, billed per token under Copilot usage-based billing. Microsoft serves it as a hosted model only (no weights), with text and image input, a 1,050,000-token context window (922,000 input and 128,000 output tokens), and selectable reasoning effort. It is not yet offered as a Microsoft Foundry API; Microsoft says OpenRouter, Vercel, Hermes, OpenCode, and Kilo will follow.
- Cloud regions
- Not tracked yet
- Lifecycle
- Current
- API capabilities
- Tool calling, structured outputs, and batch support are not tracked yet
- Prompt caching
- Published at $0.050 per million cached input tokensGitHub Copilot models and pricing
- Self-host
- Weights are not published
- Rate limits
- Not tracked yet
How to read this profile
The visual layer above carries the decisions. These notes preserve the model, ranking, coverage, and family context behind the numbers.
We track MAI Code 2, but the public leaderboard excludes this profile until enough non-generated benchmark coverage is available. Only published rows appear above.
MAI Code 2 is a proprietary model with a 1.05M context window. It uses an explicit reasoning mode, which can improve complex problem solving while adding latency and token use.
Available in GitHub Copilot since November 17, 2026 on the Pro, Pro+, Max, Business, and Enterprise plans, billed per token under Copilot usage-based billing. Microsoft serves it as a hosted model only (no weights), with text and image input, a 1,050,000-token context window (922,000 input and 128,000 output tokens), and selectable reasoning effort. It is not yet offered as a Microsoft Foundry API; Microsoft says OpenRouter, Vercel, Hermes, OpenCode, and Kilo will follow.
Official exact-value snapshot from Microsoft AI's November 17, 2026 MAI Code 2 and MAI Code 2 Flash model card. Microsoft describes MAI Code 2 as a sparse mixture-of-experts coding model with about 80B active parameters, trained from scratch for everyday enterprise coding and not yet tuned for very long-horizon agentic work. All coding rows are Microsoft's own runs on MSBench with the GitHub Copilot CLI harness (version 1.0.102, subagents disabled) at max reasoning effort, averaged over three runs per task. We map DeepSWE 1.1 (73.2%) and Terminal-Bench 2.1 (84.6%) onto existing lanes. SWE Atlas, a composite of Scale's codebase Q&A, test-writing, and refactoring sets, has no exact lane and is listed in the benchmark ledger. The announcement's cost-relative comparisons, the blind frontend preference study, and the safety and frontier-risk tables in the card stay out of scored fields. Provider-run rows keep the model unranked.
MAI Code 2 sits in the MAI-Code family with MAI-Code-1.1-Flash, MAI Code 2 Flash. 3 of 922 tracked benchmark slots currently have displayable evidence. Missing categories stay blank.
Last updated November 17, 2026. Runtime fields remain blank until a sourced snapshot exists.
Questions
How does MAI Code 2 perform overall in AI benchmarks?
MAI Code 2 has 3 source-displayable benchmark rows, but it does not qualify for a public overall rank. The available rows remain visible by category without being converted into a site-wide score. Missing evidence stays blank instead of being estimated from an earlier model.
Is MAI Code 2 good for coding and programming?
MAI Code 2 has source-displayable benchmark coverage for coding and programming, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Is MAI Code 2 good for agentic tool use and computer tasks?
MAI Code 2 has source-displayable benchmark coverage for agentic tool use and computer tasks, but the public category table does not assign it a rank there. The individual rows remain available for inspection. A missing category position means the evidence threshold was not met; it does not convert the model's unmeasured work into a zero.
Which sibling models are related to MAI Code 2?
MAI Code 2 belongs to the MAI-Code family. Related tracked variants include MAI-Code-1.1-Flash, MAI Code 2 Flash. A sibling link indicates shared lineage or a documented configuration relationship; it does not mean the variants have identical pricing, context limits, benchmark evidence, or deployment behavior. Compare before switching.
Does MAI Code 2 have full benchmark coverage on BenchLM?
No. MAI Code 2 currently has 3 source-displayable rows. The profile exposes published, non-generated evidence and leaves missing categories blank until an exact evaluation is available. Coverage describes how much was measured; it is not a penalty added to an individual benchmark result.
What is the context window size of MAI Code 2?
MAI Code 2 has a documented context window of 1.05M. That figure is the maximum combined prompt and retained-conversation space reported for this exact model; it is not the maximum output length. The profile keeps output limits separate because providers often publish those limits independently.