Microsoft Research
41 models across 13 families · Latest: Decision 1.1 (2026-11)
Advancing the state-of-the-art in AI and computing.
Microsoft Research's portfolio covers 40 active models across 13 current families, spanning coding, rag, and agents. Open a model detail page to compare provider routes and sourced benchmarks.
Covers 7 workload areas across 40 active tracked models; last verified 2026-11-18.
Use it for
- Teams evaluating coding, rag, and agents across this lab's releases
- Comparing model families before committing to a flagship
- Migration and pricing follow-ups across 40 tracked models
Do not use it for
- Choosing a hosting provider without opening a model page for price ladders
Active models
40
Current models from this lab, excluding deprecated ones
Active families
13
Current model families from this lab
Open catalog
36 open
33 open source / 3 open weights
Lowest output price
$0.070 /1M
Cheapest tracked output across active models, per 1M tokens
Latest dated release
2026-11-17
Decision 1.1
Freshness
2026-11-18
Researched 2d ago
Information
Release cadence
- Dated releases
- 5
- Latest model
- Decision 1.1
- Latest date
- 2026-11-17
Where this lab wins
- Coding: 8 tracked models with SWE-bench / HumanEval-style scores.
- RAG: 3 tracked models with ruler / needle retrieval benchmarks.
- Agentic: 2 tracked models with BFCL, tau-bench, and SWE-bench tool-use coverage.
- Long-context: 14 tracked models with context-token or InfiniteBench-class signal.
Flagship quality / price signal
- Flagship
- Phi-2
- Selection source
- Best sourced coding Q/$
- Coding grade
- A
- Benchmark / output price
- humaneval 59.7 · $0.070/1M
About
Microsoft Research is a leading research institution that drives innovation in AI and computing. Their work on machine learning, natural language processing, computer vision, and human-computer interaction has led to significant advancements in these fields. Notable contributions include the development of the Microsoft Cognitive Toolkit (CNTK) for deep learning and breakthroughs in conversational AI with the Xiaoice chatbot.
Featured models
| Model | Released | Context | Input price ($/1M) | Output price ($/1M) | License | Openness |
|---|---|---|---|---|---|---|
| Decision 1.1 | 2026-11-17 | 32K | $0.042 | $0 | Proprietary | Proprietary |
| Microsoft-Decision-1 | 2026-10-09 | 32K | $0.042 | $0 | Proprietary | Proprietary |
| Harrier OSS v1 0.6B | 2026-03-30 | 32K | - | - | MIT | Open source |
Model families
Recent releases
- Decision 1.1- 2026-11-17
- Microsoft-Decision-1- 2026-10-09
- Harrier OSS v1 0.6B- 2026-03-30
- Harrier OSS v1 270M- 2026-03-30
- Harrier OSS v1 27B- 2026-03-30
Top comparisons
- Llama 3.2 1B Instruct vs Phi-3 Mini 4k348
- Mistral Large 2 (2407) vs Phi-3 Mini 4k158
- Llama 3 8B Instruct vs Phi-3 Mini 4k146
- Llama 3 70B Instruct vs Phi-3 Mini 4k113
- Phi-4 Mini Flash Reasoning vs Qwen3.5-9B106
- Together AI - Gemma 3n-e4B vs Phi-4 Mini Flash Reasoning98
- Gemma 2 9B SahabatAI Instruct vs Phi-4 Mini Flash Reasoning86
- Doubao Seed 2.0 Pro vs Phi-4 Mini Flash Reasoning79
Explore related pages
Last reviewed: 2026-11-18. Data sourced from public lab announcements and provider documentation.










